policy.ai

Biweekly updates on artificial intelligence, emerging technology and security policy.

— Subscribe here —

Intel explores opening a domestic foundry, House Republicans announce emerging tech legislation, and Trump administration considers banning work visas
Wednesday May 13, 2020
Worth Knowing

Trump Administration Encourages Chipmakers to Build US Foundries: To decrease reliance on chipmakers in Asia, the Trump administration is encouraging Intel to build a state-of-the-art foundry in the United States, the Wall Street Journal reported. Foundries manufacture chips for third-party chip designers and governments, and many U.S. companies currently rely on foundries in Taiwan and South Korea. A domestic Intel foundry would ensure cutting-edge indigenous chip production capabilities for American chip designers and the DOD. Officials also are encouraging Taiwan Semiconductor Manufacturing Company and South Korea’s Samsung Electronics to develop state-of-the-art foundries in the United States.
Huawei Partners with Chipmakers in Preparation for US Export Controls: In anticipation of potential U.S. export controls on chips produced by Taiwan Semiconductor Manufacturing Company, Huawei is reportedly shifting some chip production toward Chinese chipmaker Semiconductor Manufacturing International Corporation and partnering with French-Italian chipmaker STMicroelectronics. A partnership with SMIC could provide Huawei with alternative chip production options if the United States blocks exports from TSMC. ST will co-design chips with Huawei, as entity list restrictions limit Huawei's access to updated software used to design chips. It remains unclear how much of Huawei’s chip production is moving to new partners.
Research Suggests Algorithmic Efficiency Is Rapidly Increasing: Researchers at OpenAI have found that algorithmic efficiency grew more quickly than Moore’s Law-driven hardware efficiency from 2012 to 2019. According to their preprint, training a neural net to a benchmark of ImageNet classification takes 44 times less computational power now than it did eight years ago. By comparison, Moore’s Law would yield only an 11-fold improvement in hardware efficiency over the same period. Policymakers should invest in measurement and assessment of AI systems to more accurately predict future changes in AI system costs, the researchers said. They are tracking the efficiency of state-of-the-art neural nets over time and calling for submissions of relevant data.
Machine Learning Spotlight — Limitations of AI COVID-19 Diagnostic Tools: Preliminary research suggests that most models using AI for COVID-19 diagnosis and prognosis are at high risk of bias and likely misleading. The researchers reviewed 31 models that use AI for online diagnosis, diagnosis based on CT scans and case prognosis prediction. According to their assessment, the reported performance of all 31 models was “probably optimistic.” Given the lack of external validation of most models, the authors recommend against relying on any current AI-based COVID-19 diagnostic tools.
Government Updates

Trump Administration Considers Banning Work-Based Visas: The Trump administration is weighing an executive order that would suspend temporary, work-based visas, according to the Wall Street Journal. In April, President Trump temporarily suspended entry for green card applicants, but some argue the proclamation did not go far enough. Last week, Sens. Cotton, Cruz, Grassley and Hawley encouraged the president to halt all new guest worker visas for at least 60 days, including H-1B, H-2B, EB-5 and the Optional Practical Training program. The exact scope of the possible restriction remains unknown. CSET researchers have argued that rolling back OPT would damage U.S. competitiveness in AI and threaten national security.

House Republicans Plan to Introduce Emerging Tech Legislative Package: On Tuesday, House Energy and Commerce Committee Republicans led by Ranking Member Walden and Consumer Protection and Commerce Subcommittee Ranking Member McMorris Rodgers previewed a 15-bill package aimed at bolstering the development of emerging technologies, such as AI, blockchain, quantum computing and autonomous vehicles. The central proposal requires the Commerce Department to study barriers to AI deployment, with the goal of reducing bureaucratic hurdles and “unleashing” private sector innovation. Other bills direct Commerce and the Federal Trade Commission to study facial recognition, and task the FTC with assessing how AI can assist in the fight against online misinformation and terrorist content. The package currently has only Republican backing.

Army to Develop Cyber Jamming Pods with Lockheed Martin: The Army has awarded Lockheed Martin $74.8 million to develop, build and test operational electronic warfare pods after the company’s successful prototype last year. The pod uses machine learning to analyze signals and make decisions without returning to base, in addition to post-mission analysis. Known as the Multi-Function Electronic Warfare-Air-Large, it will fly under the MQ-1C Grey Eagle; other variants are being explored for smaller drones and ground vehicles. The Army finalized its contract with Lockheed in January and announced it in late April.

NSCAI Co-Authors Letter on National Security Workforce: The National Security Commission on AI, the Cyberspace Solarium Commission and the National Commission on Military, National, and Public Service wrote to Senate and House Armed Services Committee leadership, encouraging them to take action on defense and national security workforce recommendations in the Fiscal Year 2021 National Defense Authorization Act. The letter recommends streamlining hiring processes and recruiting pathways, as well as increasing awareness of opportunities for digital talent to enter public service.

In Translation
CSET's translations of significant foreign language documents on AI

CSC Scholarship Study Abroad Agreement: China Scholarship Council Subsidized Study Abroad Agreement. This document is the text of an agreement between Chinese students who study abroad and the PRC Ministry of Education’s China Scholarship Council. It requires students on CSC scholarships to report regularly on the progress of their studies and mandates that those who study abroad return to China for at least two years after completing their studies overseas.

AI Standardization White Paper: Artificial Intelligence Standardization White Paper. This government-issued white paper describes China’s approach to standards setting for AI. Appendices list all of China’s existing AI standards as of January 2018, as well as those under study, and provide examples of AI applications by leading Chinese tech companies.

What’s New at CSET

REPORTSPUBLICATIONSIN THE NEWS
TSMC risks losing ground to Intel, Commerce expands export controls, and immigration proclamation worries tech companies
Wednesday April 29, 2020
Worth Knowing

TSMC Chip Plans Suggest Company May Lose Leadership to Intel: Taiwan Semiconductor Manufacturing Company may be losing ground as the most advanced manufacturing firm of its kind, according to statistics released during its latest quarterly earnings call. TSMC’s 3 nanometer chip, expected in 2023, is projected to have lower-than-anticipated transistor density. However, Intel expects to achieve comparable density in its 7 nm chip — to be released in 2021 — and greater density in its 5 nm chip, to be released in 2023. As a result, Intel could soon reclaim its status as the leading chip manufacturer in terms of transistor density.
Nvidia Purchases Mellanox for $7B: In a deal first announced in March 2019, U.S. GPU designer Nvidia completed a $7 billion purchase of Israel-based Mellanox Technologies on Monday. Mellanox provides data-center-scale networking and interconnect supplies, which facilitate running a large number of GPUs in parallel for increased computational power. After U.S. and EU regulators approved the acquisition, Chinese antitrust officials signed off last week, removing the final barrier to the deal. Mellanox will continue to operate in Israel.
MIT Ends Collaboration With iFlyTek: The Massachusetts Institute of Technology has dropped its research partnership with iFlyTek, a Chinese AI company placed on the U.S. Entity List in October for human rights violations in Xinjiang. Announced in 2018, the planned five-year collaboration between MIT’s Computer Science and Artificial Intelligence Laboratory and iFlyTek focused on computer vision, speech-to-text systems and human-computer interaction. After two of its collaborators were added to the Entity List, MIT promised to review those partnerships, and ultimately created stricter guidelines for all high-risk international proposals.
Machine Learning Spotlight — Rewriting Sentences: Microsoft announced a new tool in Word that offers sentence-level writing suggestions through machine learning. Susan Hendrich, Group Program Manager of AI and Natural Language Processing for Microsoft Office, said the system uses a Transformer-based model trained on millions of sentences. The tool allows users to give feedback so it can continue to improve. Currently available only for Word online, Microsoft plans to incorporate the system into desktop versions in the future.
Government Updates

Immigration Proclamation Concerns Tech Companies: On April 22, President Trump suspended entry for green card applicants to the United States for 60 days, with the possibility of extension. The proclamation does not affect those currently working in the country. However, some warn that it could reduce America’s access to the international tech talent pool over the long term by hurting recruiting efforts and stoking fears of tighter restrictions. CSET research has warned that a climate of uncertainty and restriction around immigration discourages AI talent from coming to the United States.

Commerce Expands Export Controls on China, Russia and Venezuela: The Department of Commerce announced Monday that it will tighten export controls for certain dual-use technologies. The update broadens the restrictions to “military end users,” makes additional technologies subject to license requirements and eliminates exemptions for civilian use. In particular, the rule expands export controls on semiconductor manufacturing equipment to Chinese companies that support the military. Nine U.S. industry associations have voiced concerns about similar changes, arguing that regulations should be narrowly tailored to avoid damaging the industry. The rules will go into effect June 29, 2020.

Bill to Create STEM Corps Introduced in the House: On April 14, Reps. Jim Banks and Andy Kim unveiled a bill to establish a STEM Corps in the Department of Defense. Under the proposed program, students in STEM fields would receive two years of tuition coverage in exchange for four years of service in the DOD; they would have the opportunity to serve the final year with an industry partner. Rep. Banks emphasized the importance of recruiting talent to work on artificial intelligence, cybersecurity and other critical projects.  

JAIC Requests Information on Testing and Evaluation: The DOD’s Joint AI Center released a Request for Information on testing and evaluation of AI systems. The RFI seeks best practices in AI testing and evaluation, software recommendations and information about relevant vendors. The JAIC wants to apply T&E processes to the Pentagon’s full range of AI and machine learning capabilities, with particular focus on human-machine interfaces, Natural Language Processing-enabled products, voice-to-text, speech-enabled products, image analysis and autonomous systems. Responses will be accepted until May 12.

In Translation
CSET's translations of significant foreign language documents on AI

Zhejiang University’s Process for Government-funded Study Abroad: Zhejiang University’s Means of Carrying Out the “2019 National Construction State-Sponsored Study Abroad Program for Graduate Students from High-Caliber Universities.” This notice describes the 2019 application process for graduate students at Zhejiang University who wished to participate in a major Chinese government-funded study abroad program. The process reflects the workings of current PRC state-sponsored study abroad programs.

What We’re Reading

Guide: Understanding AI Technology, Greg Allen, Joint AI Center (April 2020)

Report: Preliminary Findings Regarding the Department of Energy and AI, the Secretary of Energy Advisory Board Artificial Intelligence/Machine Learning Working Group (March 2020)

Report: Recommendation to Member States on the Human Rights Impacts of Algorithmic Systems, Council of Europe’s Committee of Ministers (April 2020)

Paper: AI, Human-Machine Interaction, and Autonomous Weapons: Thinking Carefully About Taking “Killer Robots” Seriously, Christopher A. Ford, Office of the Under Secretary of State for Arms Control and International Security (April 2020)

What’s New at CSET

REPORTSPUBLICATIONSIN THE NEWS
Events


What else is going on?
Suggest stories, documents to translate & upcoming events here.
Washington State Regulates Facial Recognition, Megvii Sales Drop and NSCAI Releases Recommendations
Wednesday April 15, 2020
Worth Knowing

Washington State Passes Major Facial Recognition Legislation: Governor Jay Inslee of Washington signed SB 6280 into law, making his state the first to define acceptable government use of facial recognition. Unlike other laws that ban police departments or governments from using the technology, SB 6280 allows restricted facial recognition use; its guidelines seek to provide accountability and limit surveillance. Microsoft lobbied for the legislation, and its main sponsor, state Sen. Joe Nguyen, works at the company. The law is viewed as a possible national template: California, Maryland, South Dakota and Idaho have introduced legislation with parallel language, and Microsoft has expanded its lobbying efforts to additional states.
Megvii Sales Drop After Addition to U.S. Entity List: After being added to the U.S. Entity List in October 2019, Chinese AI company Megvii saw a sharp decline in growth. In the second half of 2019, the company’s revenue grew only 2.7 percent, despite sales tripling in the first half of the year. Its IPO, originally expected to raise between $500 million and $1 billion, has been postponed. The Entity List’s prohibition on U.S. technology exports forced Megvii to pause some operations in order to assess which components of its business violated the blacklist. The company has now launched products to generate new revenue streams, including an AI-enabled temperature detection system to help fight COVID-19.
AI Field Experiences Mixed Impact of Economic Downturn: The number of tech jobs requiring AI skills globally may grow faster than expected due to the pandemic, according to research firm International Data Corporation. The firm estimates 16 percent growth in AI jobs this year — up from the 13.3 percent previously predicted for 2020 — due to anticipated new demand for AI capabilities from healthcare providers, schools and industry. But some AI startups are cutting staff, as seed funding declines and startups undergo layoffs across industries.
Machine Learning Spotlight — Understanding Glass: Researchers at DeepMind are using neural networks to predict the dynamics of glass as it changes between liquid and solid states. The team used graph neural networks to outperform physics-inspired baselines and existing AI models, achieving 64 percent accuracy in predicting particle location for longer timescales — a 40 percent improvement over previous state-of-the-art models — and 96 percent for short timescales. The researchers hope the work will serve as a testbed for applying machine learning to physical models and lead to insights about glass and other complex systems.
Government Updates

NSCAI Releases First Quarter Recommendations: The National Security Commission on Artificial Intelligence published its First Quarter Recommendations on April 1st. The commission proposed doubling non-defense R&D investment in AI to $2 billion for FY2021, strengthening the U.S. government AI workforce through a variety of programs and exemptions, improving cooperation on AI with key allies, and promoting U.S. leadership in microelectronics and 5G. The report also encouraged accelerating AI applications in the Department of Defense through a joint DOD and Office of the Director of National Intelligence steering committee.

JAIC Announces “Responsible AI Champions” for AI Ethics Principles: DOD’s Joint AI Center described its plans to ensure implementation of the DOD AI ethical principles, including launching a cross-functional team of ethics advocates. The “Responsible AI Champions” will be trained in DOD’s newly-adopted Ethical Principles for AI, and will “champion implementation practices,” educate their colleagues and identify new applications for the principles. In addition, the JAIC plans to use their Acquisition & Sustainment process to evaluate how partners are developing technologies for the DOD and has created a DOD-wide Responsible AI Subcommittee to establish a plan for implementing the ethical framework.

DOE Announces Plans to Fund $30M in AI and ML Research: The Department of Energy announced $30 million in funding for two types of research involving AI. The first category, “Scientific Machine Learning for Modeling and Simulations,” allocates $10 million in FY2020 funds for machine learning use in predictive modeling and simulation in physical sciences. In addition, $7 million will fund “AI and Decision Support for Complex Systems,” researching machine learning applications in real-time decision support for systems underpinning cybersecurity and power grid resilience. DOE intends to provide an additional $13 million for machine learning decision support research in future years. Applications for funding are open for the National Laboratories, universities, nonprofits and industry.

In Translation
CSET's translations of significant foreign language documents on AI

AI Strategic Advisory Committee: Profiles of Members of China’s New Generation AI Strategic Advisory Committee. On November 15, 2017, China’s Ministry of Science and Technology announced the formation of this committee to advise the government on S&T development plans and projects involving AI. This document summarizes Chinese government and media reporting on the committee and includes a biography of each committee member.

What’s New at CSET

REPORTSIN THE NEWS
Supercomputers used to combat COVID-19, AI as content moderator, and ML conferences go virtual
Wednesday April 1, 2020
Worth Knowing

Social Media Companies Use AI For Content Moderation: Google, Facebook and Twitter announced they are relying more heavily on artificial intelligence for content moderation instead of humans due to the coronavirus. Unlike many tech jobs that can be performed at home, content moderation is typically outsourced to contractors who are required to work from the office for privacy reasons. The companies warned that users should expect more errors from the automated systems and longer wait times until someone can investigate any appeals.
Governments Weigh Using Location Data to Combat Coronavirus: Countries around the world are debating the role of cell phone location data in monitoring the spread of COVID-19. The data can be used to track the movement of sick individuals and assess compliance with government shelter-in-place orders. However, experts are divided on what type of data collection and usage is acceptable; some recommend anonymizing and aggregating any data collected. The Wall Street Journal last week reported that the Centers for Disease Control and Prevention are already compiling aggregated location data, primarily from the advertising industry.
ML Community Experiments with Virtual Conferences: Two upcoming gatherings, the International Conference on Learning Representations and the International Conference on Machine Learning, will be held virtually on account of COVID-19, according to the organizers. ICLR reduced their registration fee to $50 for students and $100 for non-students — down from $450 and $550 respectively — and ICML announced that they are planning a similar substantial reduction. Organizers said the machine learning community had already been discussing hybrid or fully virtual conferences to combat climate change — and if these tests succeed, future conferences might shift permanently in that direction.
Government Updates

White House Makes Supercomputers Available for COVID-19 Research: Last week, the White House Office of Science and Technology Policy announced the creation of the COVID-19 High Performance Computing Consortium, which makes supercomputers available to researchers studying the coronavirus. The consortium was coordinated by the Department of Energy and IBM, and includes partners from industry, academia and federal agencies. The Director of the National Science Foundation, France Cordova, said these resources would allow scientists and engineers to pursue “data science, computational modeling and artificial intelligence approaches.” Proposals can be submitted to the consortium via their online portal.

Report on Service Recommends Investing in Tech Skills of Federal Workforce: The Final Report of the National Commission on Military, National, and Public Service recommends assisting federal agencies in hiring, developing and supporting cybersecurity, IT and STEM professionals by increasing funding and launching new programs. According to the report, the federal government has “not invested enough in maintaining and increasing” IT and cybersecurity skills in its workforce. Commissioners propose a range of recommendations to address the situation, including reskilling programs, piloting a Federal Civilian Cybersecurity Reserve, increasing benefit competitiveness and allowing for flexible hours and telework.

Proposal Advances to Restrict Exports to Chip-Makers Supplying Huawei: Cabinet officials reportedly agreed to move forward with plans to limit the export of U.S. chip-making equipment last Wednesday, according to The Wall Street Journal. The restrictions would require foreign chip-makers to obtain a license before using U.S. semiconductor manufacturing equipment to produce chips for Huawei. First considered in February, there were concerns about the limits’ potential impact on U.S. businesses, as covered in a previous edition of policy.ai.

In Translation
CSET's translations of significant foreign language documents on AI

Guidelines for China’s AI Pilot Zones: Guidelines for National New Generation AI Innovation and Development Pilot Zone Construction Work. China’s Ministry of Science and Technology issued a notice in 2019 describing a process for establishing AI innovation and development pilot zones in cities with robust AI infrastructure. China planned to create 20 AI pilot zones by 2023; so far, it has established 11 such zones, with four new ones announced in March 2020.

What We’re Reading

Special Issue: AI and Security, VentureBeat (February 2020)

Syllabus: Artificial Intelligence and China, Jeffrey Ding, Sophie-Charlotte Fischer, Brian Tse and Chris Byrd (January 2020)

Report: UK Tech for a Changing World, Tech Nation (March 2020)

Special Report: China’s Defense Budget in Context: How Under-Reporting and Differing Standards and Economies Distort the Picture, The Heritage Foundation (March 2020)

What’s New at CSET

IN THE NEWS
AI and coronavirus, TSMC weighs opening US foundry, and National AI Initiative Act introduced in the House
Wednesday March 18, 2020
Worth Knowing

AI Organizations Aid in the Fight Against Coronavirus: Over the last few weeks, AI researchers and companies have attempted to use machine learning to combat the coronavirus, but so far, the results are mostly untested. Key areas of application include:
Amid the influx of interest, experts warn against overhyping AI’s role. In addition to the newness of the technology, researchers say it’s been difficult to get reliable data on COVID-19 due to restrictions on medical information, inconsistent accounts and lack of testing. To facilitate future AI use in helping cure diseases, the Defense Innovation Board called for digitizing the world’s largest collection of human pathological tissue samples.
TSMC Weighs Opening Foundry in the US: The world’s largest contract chipmaker, Taiwan Semiconductor Manufacturing Co., may build a state-of-the-art foundry in the United States, according to the Nikkei Asian Review. The discussions follow pressure from Washington and the growing possibility of the company being caught in the U.S.-China trade war. Currently, the United States lacks a foundry capable of manufacturing the most advanced chips, so many leading-edge AI chips designed by U.S. firms are fabricated at TSMC’s Taiwan-based foundries. TSMC is purportedly considering producing chips with 2 nanometer technology in the new factory, which could make it the world’s most cutting-edge foundry.
PAI Launches Project on Publication Norms: The Partnership on AI — a multi-stakeholder group of tech companies, academics and nonprofits — is developing publication practices for responsible AI. Despite norms of openness in the machine learning community, PAI’s project explores when and how researchers should publish novel results that could be misused. PAI’s research builds on their earlier work on this topic and will involve soliciting feedback from the machine learning community, organizing a workshop and ultimately publishing a set of community recommendations.
US Home to Majority of Top AI Startups: A majority of the most promising AI startups are based in the United States, according to a new report by CB Insights. Assessing factors such as patents, investor profiles, tech novelty, market potential and team strength, the researchers created a metric to select the top 100 startups from a pool of 5,000. According to their analysis, 65 of these leading startups are located in the United States, followed by eight in both Canada and the United Kingdom and six in China. Collectively, the 100 startups have raised more than $7.4B in funding from 600 investors.
Government Updates

National AI Initiative Act Introduced in House: House Science Committee Chairwoman Johnson, Ranking Member Lucas and Reps. McNerney, Olson, Lipinski and Weber introduced the National Artificial Intelligence Initiative Act of 2020 last week. The bipartisan legislation aims to accelerate and coordinate federal investments in AI through measures including formalizing interagency coordination, creating an advisory committee and establishing AI institutes to facilitate partnerships between academia, public and private sectors. The legislation would support standards development at NIST, as well as AI research at the NSF and DOE. Sens. Heinrich, Portman and Schatz introduced a bill with many of the same aims in May, although there are several significant differences.

IC Developing AI Ethical Principles: The intelligence community is developing principles for ethical use of AI, according to Ben Huebner, Chief of the Office of the Director of National Intelligence’s Office of Civil Liberties, Privacy, and Transparency. The framework will focus on the aspects of machine learning that are new to the IC, particularly relating to transparency, privacy and accuracy. Hueber noted the ODNI has general “consensus” with the DOD’s Joint AI Center, and the guidelines will likely resemble those adopted by the DOD in late February.

Deepfakes In Federal Elections Prohibition Act Introduced in House: Reps. Lynch, DeSaulnier, Welch and Cooper introduced the Deepfakes in Federal Elections Prohibition Act, which would limit the distribution of materially deceptive audio or video prior to a federal election. The legislation would amend the Federal Election Campaign Act of 1971 to prohibit the malicious distribution of manipulated media of a candidate within 60 days of a federal election. An exception would be made if the media is labeled as manipulated.

In Translation
CSET's translations of significant foreign language documents on AI

Foreign Expert Recruitment Plan: Notice of the Office of the Ministry of Science and Technology Declaring the 2019 High-End Foreign Expert Recruitment Plan. The Ministry of Science and Technology describes a plan to entice foreign scientists, professors and entrepreneurs to work in China. Chinese companies, universities and research institutes apply on behalf of their prospective foreign employees.

AI Pilot Zones: China Creates National New Generation Artificial Intelligence Innovation and Development Pilot Zones. The Ministry of Science and Technology announced the establishment of four new AI Pilot Zones in cities across China, which encourage the expansion of China’s AI industry in cities where it is already developing. The ministry created the first seven such zones in 2019 and plans to build 20 by 2023.

What We’re Reading

Project: AI Chips, MacroPolo (March 2020)

Report: A Quantitative History of AI Research in the United States and China, Daniel Ish, Andrew Lohn and Christian Curriden (September 2019)

Report: Allies and Artificial Intelligence: Obstacles to Operations and Decision-Making, Texas National Security Review (March 2020)

Response: National Science Foundation Response to the JASON Report “Fundamental Research Security” (February 2020)

What’s New at CSET

At the request of the White House Office of Science and Technology Policy, CSET coordinated the release of the COVID-19 Open Research Dataset (CORD-19), a machine-readable dataset of scholarly literature about the coronavirus, in coordination with the Allen Institute for AI, the Chan Zuckerberg Initiative, Microsoft and the National Library of Medicine. CORD-19 was announced by the White House and covered in FedScoop, NextGov, TechCrunch and Geekwire.

Events


What else is going on?
Suggest stories, documents to translate & upcoming events here.
DOD adopts AI ethical principles, U.S. Chief Technology Officer criticizes EU approach to AI regulation, and OECD launches AI Policy Observatory
Wednesday March 4, 2020
Worth Knowing

OECD Launches AI Policy Observatory: The Organisation for Economic Co-operation and Development launched its AI Policy Observatory last Thursday. Known as OECD.AI, the new website aims to aid policymakers around the globe in developing trustworthy AI. The OECD announced plans to launch the observatory when the OECD Principles on AI were published in May 2019. The platform includes guidance on how to implement the OECD AI principles, databases of AI policies and initiatives, and AI trends and data.
EU Reportedly Considering Integrated Facial Recognition System: The European Union may create an EU-wide network of facial recognition databases, according to The Intercept. Under the Prüm Convention, the 27 EU member states currently share police fingerprint and DNA databases. The 39 countries in the Visa Waiver Program — including the United States — share similar data. However, the EU may be planning to expand the existing system to encompass police facial recognition databases from all of Europe and the United States. The development comes after the European Commission considered, but ultimately decided against, recommending a ban on facial recognition.
Machine Learning Spotlight — Identifying Antibiotics: Researchers at MIT discovered a new antibiotic using deep learning. The antibiotic, known as halicin, works against a variety of bacteria, including those thought to be antibiotic-resistant. Researchers trained the machine learning model on a dataset of existing molecules that inhibit E. coli growth, then asked it to predict which of 6,000 different molecules would prove effective against E. coli. From the results, researchers selected candidates for testing and discovered a strong antibiotic in halicin, a molecule originally identified for diabetes treatment. Experts believe this methodology could apply to a wide range of drug discoveries beyond antibiotics.
Government Updates

DOD Adopts Ethical Principles for AI: On February 24th, the Department of Defense formally approved five AI ethical principles drafted by the Defense Innovation Board. The framework covers both combat and non-combat applications of artificial intelligence, and requires AI use to be responsible, equitable, traceable, reliable and governable. However, as Director of the Joint AI Center Lt. Gen. Jack Shanahan noted, adopting the principles was “the easy part,” and the real challenge lies in understanding where to apply them. To that end, the DOD has selected lawyer Alka Patel to lead a two-person team tasked with implementing the principles.

U.S. Chief Technology Officer Criticizes EU’s Approach to AI Regulation: Chief Technology Officer of the United States Michael Kratsios critiqued the European Union’s AI regulation strategy in an interview with The Wall Street Journal. The EU should be “spending as much time on investing in these technologies as they are in drafting white papers,” he stated, characterizing the EU’s investment in AI as “very small.” Kratsios said the recently released European Commission’s White Paper on AI “clumsily attempts” to categorize all AI technology as either high or low risk; he called instead for a more nuanced spectrum.

White House Releases First Annual Report on AI: The Office of Science and Technology Policy published its American Artificial Intelligence Initiative: Year One Annual Report, an overview of accomplishments since President Trump launched the American AI Initiative. The document summarizes progress to date as well as long-term vision, including increasing AI R&D, facilitating access to AI resources such as data and compute, and drafting AI regulatory principles to remove barriers to innovation.

In Translation
CSET's translations of significant foreign language documents on AI

Members of the PRC Ministry of Education AI Expert Group: Notice of the Office of the Ministry of Education Regarding the Establishment of the Ministry of Education AI Technology Innovation Expert Group. Announcement of the formation of the AI Technology Innovation Expert Group in 2018, tasked with offering higher education institutions advice on AI talent cultivation, innovation, and cooperation between industrial and academic institutions. The translation has been annotated with information about each member of the group.

What We’re Reading

Principles: Rome Call for AI Ethics, The Vatican (February 2020)

Report: A 20-Year Community Roadmap for AI Research in the US, Computing Community Consortium (August 2019)

Opinion: America Must Shape the World’s AI Norms — or Dictators Will, William Cohen, Leon E. Panetta, Chuck Hagel and Ash Carter (February 2020)

What’s New at CSET

REPORTSIN THE NEWS
Events


What else is going on?
Suggest stories, documents to translate & upcoming events here.
The White House AI budget, Moscow's facial recognition system, and the Ethical Use of Facial Recognition Act
Wednesday February 19, 2020
Worth Knowing

Moscow Expands Its Facial Recognition System: Moscow’s facial recognition system has been fully operational since January 1st. The system cost $53.3M and consists of 175,000 cameras, making Moscow one of the world’s most surveilled cities. Although the technology has been in development since 2017, Russian facial recognition company NtechLab recently won a contract to provide video detection services, allowing the implementation of facial recognition components. The system’s rollout has met with some resistance, including a court challenge and an activist campaign urging people to resist the technology by painting their faces.
AAAI-20 and AIES Conferences Held In New York: The 34th Association for the Advancement of Artificial Intelligence Conference on AI was held February 7-12 in New York. With more than 3,000 attendees, AAAI is one of the largest annual AI conferences, though organizers anticipated decreased attendance by Chinese scholars because of the coronavirus. Paper submissions reached a record 8,800; last year’s total was 7,700. For the third year in a row, the conference was co-located with a conference on Artificial Intelligence, Ethics, and Society
Government Updates

White House Proposed Budget Increases AI Spending, Cuts Overall R&D: The President’s 2021 budget proposes significant increases to AI spending and commits to doubling non-defense AI R&D by 2022. Relative to last year’s request, the budget proposes:
  • Increasing the JAIC budget by $48M
  • Increasing NSF AI R&D by $376M
At the same time, it also requests broad cuts to research, which some researchers say could harm AI. Relative to current enacted funding levels for 2020, the budget would:
  • Decrease NSF funding overall by $540M
  • Decrease aggregate federal funding for basic research by $2.8B and for applied research by $5.1B
  • Decrease NIST funding by $296M
The President’s budget request has included similar research cuts in previous years, but Congress has rejected them.

Senators Introduce Facial Recognition Bill: Senators Merkley and Booker introduced legislation that, if enacted, would place a moratorium on the federal use of facial recognition pending further congressional action. The Ethical Use of Facial Recognition Act establishes a 13-member commission tasked with recommending guidelines and limitations for the technology, and lays out a process for considering legislation. Until then, federal use of facial recognition would be suspended, though police could still use it with a warrant. There is currently no companion bill in the House.

Judge Blocks JEDI Contract: The U.S. Court of Federal Claims blocked the Pentagon from moving forward with its $10 billion cloud computing contract. Last Thursday, Judge Patricia E. Campbell-Smith granted Amazon’s motion for a preliminary injunction against the DOD’s Joint Enterprise Defense Infrastructure project contract. JEDI is a 10-year contract to modernize the tech infrastructure of the DOD, including implementing AI for defense operations. Microsoft was awarded the contract in October, but Amazon argued that Trump’s “personal animus” toward the company biased the selection process. The work, scheduled to begin last Friday, is now suspended until further notice.

Commerce Reportedly Considers Restricting Exports to Chip-Makers Supplying Huawei: The Department of Commerce is reportedly considering new limits on the export of U.S. chip-making equipment. The restrictions would require foreign chip-makers to obtain a license before using U.S. semiconductor manufacturing equipment to make chips for Huawei. However, some experts warn this regulation would only hurt U.S. companies if done unilaterally, as foreign chip-makers could obtain the equipment from other countries. On Tuesday, though, President Trump tweeted that sales of U.S. technology to China should not be limited, making the likelihood of restrictions unclear.

In Translation
CSET's translations of significant foreign language documents on AI

Foreign Expert Talent Notice: Notice on Applying for 2020 National Foreign Expert Projects. Description of three Chinese government programs designed to recruit foreign talent in 2020, particularly in tech fields. The notice explains the purpose of each talent program and the benefits conferred upon foreign experts who are accepted.

What We’re Reading

Paper: University of Pennsylvania Input to National Security Commission on Artificial Intelligence, Penn Global and Perry World House (February 2020)

Report: Artificial Intelligence and Public Standards, UK Committee on Standards in Public Life (February 2020)

Report: A Brief Examination of Chinese Government Expenditures on AI R&D, Institute for Defense Analyses (February 2020)

What’s New at CSET

PUBLICATIONSIN THE NEWS
EU AI white paper leaked, global AI startup funding rises and Shanahan to retire
Wednesday, February 5, 2020
Worth Knowing

Leaked European Commission White Paper Proposes AI Regulation: A leaked white paper allegedly from the European Commission explores possible methods for AI regulation, ranging from legally binding requirements for “high-risk” AI to an optional “trustworthy AI” label. The document also considers a ban on facial recognition, but later reporting suggests that proposal has been dropped. Newly appointed Commission President Ursula von der Leyen has promised to propose legislation governing AI within her first 100 days in office. The final European Commission plan is expected on February 19th.
Global Funding for AI Startups Rises, Chinese Funding Declines: Funding for AI startups globally rose to a record $27 billion in 2019, according to a new CB Insights report. The number represents a significant increase over the amounts raised globally in 2018 ($22.1 billion) and 2017 ($16.8 billion). Of the $27 billion raised, U.S. startups accounted for $17 billion, up from $13.3 billion the previous year. Chinese AI startups, on the other hand, raised only $2.9 billion in 2019, down from $4.7 billion in 2018. However, the number of deals closed in China continued to rise. 
Google Announces New AI Chatbot and Benchmark: Last week, Google AI unveiled Meena, a neural net chatbot capable of more relevant conversations than previous state-of-the-art chatbots. While it’s an improvement over previous chatbots, the researchers say one of their main contributions is a benchmark for assessing the relevance of chatbot comments, known as the Sensibleness and Specificity Average. According to their calculations, Meena outperforms existing chatbots with a 79 percent SSA, as compared to 56 percent in previous models. Humans, by contrast, achieved an 86 percent SSA. The researchers also found an “automatic metric” — one that doesn’t require human judgment — in the neural model that is highly correlated with SSA. If accurate, it suggests a potential avenue for improving this component of chatbots.
Facebook Announces Near-Perfect Point-Goal Navigation: Facebook AI Research created a model, DD-PPO, that has “effectively solved” the task of point-goal navigation. In point-goal navigation, an agent must navigate from a random starting location to a specific destination in an unfamiliar environment without a map. Using only a camera, GPS and compass, DD-PPO achieved 99.9 percent success, relative to previous systems at 92 percent success. The model was trained using reinforcement learning and tested in simulated and real-world environments using the open-source robot LoCoBot. DD-PPO could have applications in both civilian and military navigation systems.
Government Updates

JAIC Director Shanahan Retiring: Director of the Defense Department’s Joint AI Center Lt. Gen. Jack Shanahan will retire this summer after more than 35 years of service in the U.S. Air Force. Shanahan has led the JAIC since its creation in 2018 and previously oversaw Project Maven. In addition to leading in AI, he advocated for other emerging technologies, including cloud computing. The search for his replacement is ongoing.

DOJ Charges Three Academics in China-Related Cases: On January 28th, Harvard University Chemistry and Chemical Biology Department Chair Dr. Charles Lieber was arrested and charged with lying about his involvement in a Chinese talent recruitment program and illegally hiding payments of more than $1.5 million from the Chinese government. On the same day, Chinese national Yanqing Ye, a researcher at Boston University, and Zaosong Zheng, a researcher at Harvard University, were also charged with crimes related to aiding the Chinese government. The arrests come amid greater FBI scrutiny of China’s involvement with U.S. research institutions.

Bennet Criticizes White House Guidance for AI Regulation: Senator Bennet sent a letter to U.S. Chief Technology Officer Michael Kratsios strongly critiquing the White House guidance for AI regulation released last month. In the letter, he calls the document “little more than gauzy generalities” that fail to address coordination with allies, privacy protections or civil rights. The administration’s AI-related efforts “lack the long-term vision and resources” to solidify U.S. competitiveness, Bennet says.

In Translation
CSET's translations of significant foreign language documents on AI

Chinese Media Reaction Brief: Chinese Reactions to Fall 2019 U.S. AI-Related Initiatives. Translations of Chinese media reactions to three U.S. AI-related policy initiatives in the fall of 2019: the addition of 28 Chinese organizations to the Entity List, the Defense Innovation Board AI Principles and the National Security Commission on AI Interim Report.

What We’re Reading

Issue: Strategic Competition for Emerging Military Technologies: Comparative Paths and Patterns, Michael Raska, National Defense University Press (January 2020)

Report: Rising to the China Challenge, CNAS (January 2020)

Special Issue: Emerging Technologies, Strategic Trade Review (Winter / Spring 2020)

Commentary: An Alternative to the Defense Department’s New, Technology-Focused Organizations, Morgan Dwyer (January 2020)

What’s New at CSET

REPORTSPUBLICATIONSIN THE NEWS
Facial recognition regulations evolve, Industries of the Future Act introduced and trade deal with China tackles tech transfer
Wednesday, January 22, 2020
Worth Knowing

Facial Recognition Controversy Deepens: More than 600 U.S. law enforcement agencies have quietly begun using a facial recognition app from start-up Clearview AI, finds a New York Times investigation. The system, which scrapes images from Facebook, YouTube, Venmo and others, has amassed more than three billion photos. Clearview’s volume of photos and lack of regulation or independent testing have exacerbated existing concerns about the increasing use of facial recognition. Meanwhile, the patchwork of state and local regulations on facial recognition continues to evolve: Last week, Cambridge became the fourth city in Massachusetts to limit the technology, and the California law prohibiting facial recognition use by law enforcement went into effect on January 1.
AI Featured at 53rd Annual CES: Nearly 170,000 people attended the 2020 International Consumer Electronics Show in Las Vegas earlier this month. Much of the discussion centered on AI, including:
In addition, exhibitors unveiled a large variety of AI applications, ranging from AI avatars to autonomous navigation systems to virtual assistants.
DeepMind Publishes Protein Folding Results: DeepMind published a study on AlphaFold, their protein folding predicting system, in top science journal Nature last week. First posed in 1962, the protein folding problem asks how a given chain of amino acids folds into the 3D structure of a protein. Answering this problem is an important step in understanding the biochemistry of living organisms. DeepMind’s discovery came in December 2018 when AlphaFold won CASP, the biennial protein prediction competition, by correctly predicting 24 out of 43 structures; the runner-up predicted only 14. AlphaFold’s progress demonstrates the potential for scientific advancement through machine learning.
Government Updates

Senators Introduce Bill to Support R&D for AI and Other Industries: Sens. Wicker, Gardner and Baldwin introduced the Industries of the Future Act last week. The bipartisan legislation would require a plan to increase federal investments in “industries of the future” — including AI, quantum computing and biotechnology — to $10B per year by 2025. It would also establish an Industries of the Future Coordination Council to advise the White House Office of Science and Technology Policy on federal measures necessary to maintain the U.S. global edge in emerging technologies.

Trade Deal With China Prompts Debate Over Tech Provisions: President Trump signed a “Phase One” trade agreement with China on January 15 that included sections on intellectual property theft and technology transfer. China agreed to stop requiring U.S. companies to transfer technology as a condition of operating in the country. It also committed to strengthening legal protections for American intellectual property — including harsher punishments for IP theft — and improving the criminal and civil procedures for combatting online patent and copyright infringement. However, the agreement has triggered debate over the likelihood of implementation and enforcement, while prospects for a "Phase Two" deal remain uncertain.

House Holds Hearing on Facial Recognition: On January 15, the House Committee on Oversight and Reform held its third hearing on facial recognition technology. Chairwoman Maloney indicated that the committee plans to introduce “common-sense” facial recognition legislation in the very near future. Despite its expanded private-sector use, facial recognition is “just not ready for prime time,” she said. This hearing follows a National Institute of Standards and Technology report finding that commercial facial recognition systems misidentify women and minorities at high rates. Ranking Member Jordan also committed to advancing a bipartisan bill.

In Translation
CSET's translations of significant foreign language documents on AI

China’s Strategy for Science and Technology Innovation: National 13th Five-Year Plan for S&T Innovation: Translation of a PRC State Council plan for science and technology innovation from 2016 to 2020. The first half of the plan details specific technologies that are near term priorities for research and investment. The second half discusses proposed changes to China’s S&T infrastructure.

WAIC Proposed Guidelines for AI Security: World AI Conference Security and Rule of Law Guidelines: Translation of a document issued just before the World Artificial Intelligence Conference in Shanghai in August 2019. The document consists of proposed legal guidelines to address a wide range of potential dangers posed by the rise of AI technology, including bugs, hackers, algorithmic bias and unemployment.

What We’re Reading

Strategy: Artificial Intelligence in Support of Defense (available in English and French), Report of the AI Task Force of the French Ministry of the Armies (September 2019)

Report: The Global AI Index, Tortoise Media (December 2019)

Article: The Offense-Defense Balance of Scientific Knowledge: Does Publishing AI Research Reduce Misuse?, Toby Shevlane, Allan Dafoe (December 2019)

What’s New at CSET

PUBLICATIONSIN THE NEWS
Events


What else is going on?
Suggest stories, documents to translate & upcoming events here.
JAIC funded at $184M, U.S. export restrictions expanded and Facebook bans deepfake videos
Wednesday, January 8, 2020
Worth Knowing

2019 AI Index Documents the Field’s Progress: Stanford University’s Institute for Human-Centered AI released its annual AI Index Report. The almost 300-page report tracks the development of AI over time across a broad range of dimensions. Among the findings: significant growth over the past few years in AI conference attendance, number of publications, investment levels, and enrollment in related education. While China led in some metrics, including total number of publications, the United States led in others, such as citation impact, investments and patents. The report also includes a Global AI Vibrancy Tool that lets users compare countries’ relative strengths in AI.
OECD: Chinese Semiconductor Firms Benefit Most From Government Support: The Organization for Economic Co-operation and Development published a report on government support in the semiconductor industry, which found that total government assistance (including grants, tax concessions, below-market debt and below-market equity) to private semiconductor companies from 2014 to 2018 topped $50 billion worldwide. Chinese companies received the most support, amounting to 85 percent of all below-market equity and 98 percent of all below-market debt that the study identified. For SMIC and Tsinghua Unigroup in particular, government support made up more than 30 percent of their annual revenue. 
Facebook Curtails Artificially Generated Media: Facebook announced Monday that it is banning deepfake videos, specifically those produced by artificial intelligence to appear authentic, and created or edited to mislead. Critics of the policy noted that it would not limit so-called “cheapfakes,” videos manipulated without use of artificial intelligence. In late December, Facebook also removed several hundred fake accounts that used artificially generated photos as part of a coordinated disinformation campaign linked to Epoch Media Group. The photos, identified as products of a publicly available model called StyleGAN, marked the first time Facebook has identified systemic inauthentic use of AI-generated photos.
Government Updates

Commerce Department Restricts Export of Certain AI Software: The Bureau of Industry and Security amended the Export Administration Regulations on Monday to include restrictions on the export of geospatial AI software. The interim rule requires a license for export and reexport of this software to all destinations except Canada. Restricted software must use a deep convolutional neural network to automate the analysis of geospatial imagery and have a variety of specific characteristics. The rule is open for comment until March 6th. Additional restrictions on exports of emerging technologies are expected.

FY20 Appropriations Increase Funding for AI: The Fiscal Year 2020 appropriations act passed by Congress and signed by the President in late December included substantial investments in AI-related activities. It funded the Joint AI Center at $183.83 million. While $25 million below the President’s request, this was a significant increase over the FY2019 funding of $93 million. Overall, the measure provided $77.5 million above the President’s request for Department of Defense AI-related activities.

NDAA-Mandated RAND Report Finds DOD Unprepared to Integrate AI: A federally mandated report on the Defense Department’s posture in AI found that the DOD’s approach is “significantly challenged across all dimensions.” The RAND Corporation’s independent assessment concluded that the JAIC lacks the authority and resources to implement the DOD’s vision for AI. In addition, the authors determined the current state of verification, validation, test and evaluation is “nowhere close” to ensuring the safety of AI applications. The report recommends new governance structures and strategic planning initiatives, among other actions.

White House Proposes Guidance for AI Regulation: On Tuesday, the Office of Management and Budget proposed guidance for government agency regulation of AI in the private sector. Under Executive Order 13859 on Maintaining American Leadership in Artificial Intelligence, agencies with regulatory authority must submit AI regulation plans to OMB. The memorandum outlines principles the plans should follow, including promoting trustworthy AI without hampering innovation and growth. It also encourages the support of voluntary consensus standards developed by industry for self-regulation. Agency plans are due in 180 days. Chief Technology Officer of the United States Michael Kratsios wrote an op-ed in Bloomberg introducing the principles.

What We’re Reading

Report: The American AI Century: A Blueprint for Action, CNAS (December 2019)

Report: 2019 AI Now Report, AI Now (December 2019)

Report: Report on Artificial Intelligence: Implications for NATO’s Armed Forces, NATO Parliamentary Assembly Science and Technology Subcommittee (October 2019)

Paper: A Stable Nuclear Future? The Impact of Autonomous Systems and Artificial Intelligence, Michael C. Horowitz, Paul Scharre and Alexander Velez-Green (December 2019)

In Translation
CSET's translations of significant foreign language documents on AI

Qianzhan 2019 China AI Industry Report: 2019 Report on Current Conditions and Trends in the Artificial Intelligence Industry: Translation of Qianzhan Industry Research Institute’s business analysis of China’s AI industry. The document analyzes the current supply chain, market development and investments in China’s AI industry. It also assesses the outlook and trends for the future of the industry.

China’s Strategy for Innovation-Driven Development: Outline of the National Innovation-Driven Development Strategy: Translation of a CPC Central Committee and PRC State Council strategy identifying industries that China feels would most benefit from increased indigenous innovation. The document also identifies foreign talent and technology transfer as crucial for China’s emerging technology sectors.

China’s Ten-Year Strategy for Education Reform: Outline of the National Plan for Medium- and Long-Term Education Reform and Development: Translation of a CPC Central Committee and PRC State Council strategy for education reform issued in July 2010. Although the strategy doesn’t mention emerging technologies explicitly, the document addresses international educational exchange and cultivation of world-class talent, which has implications for emerging technology.

What’s New at CSET

REPORTSPUBLICATIONSIN THE NEWS
Events


What else is going on?
Suggest stories, documents to translate & upcoming events here.
Chinese Public AI R&D Spending, NeurIPS Underway, and NDAA Includes Provisions on AI
Wednesday December 11, 2019
Worth Knowing

NeurIPS Underway in Vancouver: The Neural Information Processing Systems Conference, the most attended annual AI conference, is in progress in Vancouver, Canada until December 14. Conference organizers anticipated more than 13,000 attendees — a significant increase from the 8,000 participants in 2018. A total of 9,185 papers were submitted for consideration, with 1,428 accepted for presentation. NeurIPS is slated to be held in Vancouver again in 2020 and Sydney, Australia in 2021.
China Legislates Deepfakes: The Cyberspace Administration of China will require that all deepfakes be clearly marked as artificially generated starting January 1, 2020. The rules apply to all “fake news” created with technologies such as artificial intelligence or virtual reality. Failure to comply will be a criminal offense. Officials cite threats that deepfakes pose to national security and the social order as motivating factors. While there is no comparable deepfake legislation in the United States, California has passed laws restricting deepfake use under specific circumstances.
Reinforcement Learning Vulnerable to Attacks During Training: Reinforcement learning algorithms can be sabotaged by subtle tweaks in training data, according to new reporting from Wired. Known as Trojan attacks, these attacks alter the data used to train the machine learning system and result in specific undesired behavior when the system is implemented at a later date. Researchers at Boston University found that they could complete the attack by altering just .025 percent of the training data, and successfully demonstrated this discovery on a DeepMind algorithm. The researchers believe theirs is the first demonstration of Trojan attacks on reinforcement learning agents, which learn from their environment.
Government Updates

NDAA Extends NSCAI Mandate, Enhances Hiring for JAIC: House and Senate negotiators have reached an agreement on the Fiscal Year 2020 National Defense Authorization Act. The conference report incorporates several provisions related to AI, including authorization for the Joint AI Center to enhance its hiring of science and engineering experts. The NDAA also extends the National Security Commission on AI’s mandate until October 2021, requires a second interim report by December 2020 and delays the date of the final report until March 2021. In addition, the NDAA directs the Department of Defense to provide an analysis comparing U.S. and Chinese capabilities in AI and to report on the JAIC’s mission and objectives.

Schmidt and Work: US in Danger of Losing Global Leadership in AI: In an op-ed published last week, the co-chairs of the National Security Commission on AI, Eric Schmidt and Bob Work, wrote that the United States must act quickly to avoid losing its technical lead to China. While the country has long been a world leader in AI, they warn that by many metrics, America’s lead is dwindling. The op-ed summarizes the findings of the NSCAI Interim Report and underscores the importance of AI to national security and economic prosperity.

ICIG Report Describes Activities to Improve Oversight of AI: The Office of the Inspector General of the Intelligence Community released its Semiannual Report detailing its goals and activities from April to September 2019. One of the ICIG’s five programmatic objectives in 2019 was improving oversight of artificial intelligence. To that end, the report describes steps the ICIG took to build collaboration around and understanding of AI, both within and outside the intelligence community. The report also discusses the possibility of building an ICIG Community of Interest on AI.

What We’re Reading

Special Issue: RUSI Journal on Artificial Intelligence, The Royal United Services Institute (November 2019)

Post: An Epidemic of AI Misinformation, Gary Marcus in The Gradient (November 2019)

Paper: Review of Dual-Use Export Controls, European Parliament Think Tank (November 2019)

In Translation
CSET's translations of significant foreign language documents on AI

China’s Five-Year Industrial Strategy for Emerging Technology: Circular of the State Council on Issuing the National 13th Five-Year Plan for the Development of Strategic Emerging Industries: Translation of a PRC State Council plan that sets quantifiable goalposts for the growth of certain high-tech industries. An appendix specifies the Chinese ministries responsible for carrying out this plan for each type of emerging technology.

What’s New at CSET

REPORTSIN THE NEWS
Cerebras unveils the CS-1, explaining NLP, and China’s plan for university-level AI education
Wednesday November 27, 2019
Worth Knowing

Cerebras Unveils Computer Using Largest Chip Ever Built: Last week, Cerebras Systems announced the CS-1, a computer designed around its Wafer Scale Engine. The WSE, released in August, is specialized for deep learning. It is also the largest computer chip ever produced, at more than 50 times larger than standard chips. The CS-1 provides the infrastructure for users to work with the chip, beginning with Argonne National Laboratory, the company’s first partner. Cerebras says the CS-1 delivers the performance of 1,000 GPUs combined, though this claim has not been verified.
Toolkit for Explaining Natural Language Processing Predictions: Researchers at the Allen Institute for AI created “AllenNLP Interpret,” a set of tools to help explain results from natural language processing models. The software can identify which words in a given input have the most influence over a model’s prediction, and identify hypothetical alterations that would result in different predictions. The demo uses saliency maps to display words most important for the model’s output. The authors hope AllenNLP Interpret will help researchers better understand the models they develop.
Hikvision Markets AI Camera that Identifies Uyghurs: Hikvision, the world’s largest video surveillance product manufacturer, marketed an AI camera that can automatically distinguish Uyghurs from Han, surveillance information platform IPVM reported. When contacted by IPVM, Hikvision removed the product page from its website and refused to comment. IPVM’s Charles Rollet wrote that Hikvision’s rapid removal of the item suggests other companies engage in similar identification but self-censor. This is the second time in two years that the Chinese company has been confronted by IPVM about minority analytics and removed information from its website.
What Natural Language Processing Means for Disinformation: The rapid progress in natural language processing poses both challenges and opportunities for combatting disinformation. Researchers at Middlebury argue that the availability of NLP models like GPT-2 will significantly increase the risk of fake content. At the same time, cybersecurity company FireEye is exploring how to combat information operations with natural language processing models. The company claims it can use neural networks to detect patterns within text that suggest it was generated by deep learning-based language generators.
Government Updates

2019 Annual USCC Report Highlights Emerging Technology: The U.S.-China Economic and Security Review Commission submitted its annual report to Congress on the national security implications of the economic relationship between the United States and China. The report includes a section on emerging technologies and military-civil fusion that argues Chinese advancements in AI could undermine U.S. economic and military advantages. The Commission makes several recommendations to Congress, including reestablishing a higher education advisory board under the FBI to identify signs of technology transfer. However, critics have noted errors and hyperbole in the report regarding China’s space program.

2016–2019 Progress Report Published on Advancing AI R&D: The National Science and Technology Council has released its Progress Report on AI R&D. The report describes how federal agencies are advancing the field in accordance with the National AI R&D Strategic Plan. It divides AI research by national strategy, sector and agency contribution, emphasizing the breadth and depth of federal investments in AI.

What We’re Reading

Report: Characteristics of H-1B Specialty Occupation Workers: Fiscal Year 2018 Annual Report to Congress, U.S. Citizenship and Immigration Services, Department of Homeland Security (November 2019)

Strategy: National Artificial Intelligence Strategy: Advancing Our Smart Nation Journey, Smart Nation Singapore (November 2019)

Paper: Artificial Intelligence in Land Forces: A Position Paper, The German Army Concepts and Capabilities Development Centre, Bundeswehr (October 2019)

In Translation
CSET's translations of significant foreign language documents on AI

China’s Plan to Improve University-Level AI Education: The Artificial Intelligence Innovation Action Plan for Institutions of Higher Education: Translation of a Ministry of Education plan issued in April 2018. The plan lays out objectives designed to significantly enhance China’s cadre of AI talent and its university AI curricula by 2030.

China’s Plan to Build a National Tech Transfer System: The Program to Build a National Technology Transfer System: Translation of a PRC State Council plan issued in 2017. It briefly addresses China’s system for acquiring foreign technology but primarily focuses on the transfer of technology within China.

What’s New at CSET

PUBLICATIONSIN THE NEWS
  • Hewlett Foundation: CSET was awarded a $2 million grant to support a new Cybersecurity and AI project led by Ben Buchanan. CyberAI will explore the effects of automation on cyber offense and defense.
  • Syracuse University News: As part of a new $500,000 partnership with CSET, Syracuse University Institute for Security Policy and Law will assist CSET in investigating the legal, policy and security impacts of emerging technology. Judge James Baker is the grant’s principal investigator.
  • South China Morning Post: Helen Toner spoke about China’s AI ambitions in an article about Chinese reliance on U.S. technology.
  • CNAS: Andrew Imbrie and Elsa Kania have become members of CNAS’s newly launched Digital Freedom Forum.
CITATIONS
Events


What else is going on?
Suggest stories, documents to translate & upcoming events here.
China’s emotion recognition system, the DOD AI ethics recommendations and the NSCAI report
Wednesday, November 13, 2019
Worth Knowing

China Expands ‘Emotion Recognition’ AI Despite Expert Skepticism: Emotion recognition systems generated excitement at China’s 2019 Public Security Expo, the Financial Times reported. The technology is being rolled out in Xinjiang as part of crime prediction systems along with facial recognition, gait recognition and eye tracking. The system is meant to predict violent behavior by using AI to identify signs of aggression, nervousness and stress. However, experts have pushed back on such characterizations with studies suggesting that emotion recognition is accurate only 20–30 percent of the time. Despite this, both Chinese companies and U.S. tech giants like Google, Amazon and Microsoft continue to develop emotion recognition systems.
Canada Again Denies Travel Visas to African AI Researchers Attending NeurIPS: For the second year in a row, at least 15 AI researchers and students were denied travel visas to Canada over concerns that they would not return home after attending NeurIPS — a leading AI research conference — and the Black in AI workshop. Last year, Canada denied almost 100 African researchers’ visas for the same reason. ICLR, another top AI conference, will be held in Ethiopia in 2020 in part to avoid similar visa issues.
Reports of U.S. Military’s Extensive Facial Recognition System: The U.S. military maintains an extensive biometrics and facial recognition system with more than 7.4 million identities, according to documents obtained by OneZero. The system, known as the Automated Biometric Information System, stores biometric information on anyone who comes into contact with U.S. military systems abroad, including allied soldiers, with the goal of “denying... adversaries anonymity.” ABIS currently links to state and local law enforcement biometric systems and may eventually integrate with the Department of Homeland Security’s biometric database. While much is still unknown about the system, the use of facial recognition raises privacy and bias concerns given facial recognition’s low accuracy rates for women and minorities.
DeepMind Reaches Grandmaster Level in StarCraft II: DeepMind announced that AlphaStar, its AI designed to play StarCraft II, recently outperformed 99.8% of human players. Unlike the previous version of AlphaStar announced in January, changes were made to level the playing field with humans: AlphaStar now “sees” the board through a camera, is limited to act at human speed and is able to play as any species. The results are comparable to those of OpenAI’s Dota 2 artificial agent, showing the power of reinforcement learning as a training mechanism for complex gameplay environments.
Government Updates

DIB Approves AI Ethics Principles: After a series of roundtables and discussion, the Defense Innovation Board voted unanimously on October 31st to recommend AI Ethics Principles for the DOD, accompanied by a supporting document. The principles, intended for both combat and non-combat systems, state that AI should be responsible, equitable, traceable, reliable and governable. They also recommend a series of next steps, including establishing a DOD-wide steering committee and formalizing the principles within the DOD.

NSCAI Submits Interim Report to Congress: On Nov 4th, the bipartisan National Security Commission on Artificial Intelligence released its interim report. The report assesses the challenges and opportunities AI poses for national security, as well as noting concerning trendlines relative to China. The Commission suggests accelerating public investment in AI R&D, applying AI to national security, training and recruiting AI talent, protecting the U.S. technological advantage and encouraging global cooperation. Full recommendations will be made to Congress in a later report.

NSCAI Holds First-Ever Conference: The National Security Commission on Artificial Intelligence held a conference on the future of AI and national security on November 5th. Notable remarks include:
For more information, check out the videos of the second half of the conference, capped by Jason Matheny interviewing White House CTO Michael Kratsios.

What We’re Reading

Report: Report of Estonia’s AI Taskforce, published by the Republic of Estonia Government Office and the Republic of Estonia Ministry of Economic Affairs and Communications (May 2019)

Report: The Impact of Artificial Intelligence on Strategic Stability and Nuclear Risk, Volume II, East Asian Perspectives, published by the Stockholm International Peace Research Institute and edited by Lora Saalman (October 2019)

Report: Partial Disengagement: A New U.S. Strategy for Economic Competition with China, by Charles W. Boustany and Aaron L. Friedberg for the National Bureau of Asian Research (November 2019)

In Translation
CSET's translations of significant foreign language documents on AI

Taiwan’s Sensitive Science and Technology Protection Bill: General Notes on the Sensitive Science and Technology Protection Bill: Translation of a bill proposed in Taiwan’s parliament that provides for up to seven years in prison or a $1 million fine for leaks of sensitive technology. The bill aims to counter Chinese industrial espionage and reassure U.S. firms that they can conduct R&D in Taiwan without fear of their proprietary technology being disclosed to Chinese competitors.

What’s New at CSET

PUBLICATIONSIN THE NEWS
Events


What else is going on?
Suggest stories, documents to translate & upcoming events here.
Russia’s AI strategy translated, Chinese AI supply chains and the Deepfake Report Act
Wednesday, October 30, 2019
Worth Knowing

Chinese companies look to secure supply chains after being added to Entity List: Chinese AI companies seek to adapt their hardware supply chains after being put on the Entity List, which prohibits them from purchasing certain U.S. technologies. The CEO of Chinese AI startup Megvii says the company will be restricted in its ability to purchase x86 servers, GPUs and CPUs, but it will move forward with its IPO as planned. Also of note: The New York Times reports that in light of rising tensions with China, the Pentagon has been meeting with tech companies to assess U.S. dependence on Taiwanese chips, which are crucial for military applications.
Researchers use machine learning to develop a new metamaterial: A research team at Delft University of Technology created a new super-compressible material with the help of machine learning. While testing new materials usually requires extensive trial and error, the use of AI allowed for experimentation solely via simulation, significantly accelerating the process. Lead author Miguel Bessa says while the new material is exciting, the role of machine learning in its development is the real accomplishment. The researchers also released their code to facilitate broader use of ML in future materials design.
Inaugural Turing AI Fellows class named as part of UK talent push: The Alan Turing Institute, the UK Office for AI, and UK Research and Innovation have announced the appointment of five Turing AI Fellows, senior AI researchers selected to receive significant funding for five years. The institutions also published a call for applications to the Turing AI Acceleration Fellowship and Turing AI World-Leading Researcher Fellowships, which together will receive 37.5 million pounds ($48.2 million) in funding. These initiatives are part of a broader UK government strategy to attract and retain top AI talent.
Data labeling market expected to grow dramatically: The data labeling industry is growing, with workers in developing countries generating the massive quantities of human-labeled data needed to train AI. The market for data labeling was estimated at $150 million in 2018 and is predicted to grow to $1 billion by 2023. Labeled data is essential for supervised learning, and the growing industry allows tech companies to outsource this work rather than do it in-house. While changes in AI training methods could eventually make the industry less essential, it’s a necessity for now.
Government Updates

Trump reestablishes science and technology advisory council: On October 22nd, President Trump issued an Executive Order reconstituting the President’s Council of Advisors on Science and Technology and appointed the first seven members of an eventual 17. PCAST will advise the White House on science and technology and respond to requests for analysis or advice. Several of the advisors have backgrounds in artificial intelligence, which the Executive Order specifically mentioned as a key emerging technology along with quantum computing.

Senate passes Deepfake Report Act: The Senate passed the Deepfake Report Act by unanimous consent on October 24th. The bill would require the Secretary of Homeland Security to publish an annual report on the state of deepfake technology. The report is to include an assessment of technologies, how deepfakes could be used by foreign governments and non-state actors, methods for deepfake detection and progress on technological countermeasures. A companion bill was introduced in the House in June, but has not yet been brought up for a vote.

U.S. Army announces plans to integrate and adopt AI: Earlier this month, the Army provided a series of updates on its use of AI as part of its strategy for seamless AI integration. Among the developments: efforts to create an AI assistant for tank warfare known as Project Quarterback, an AI system designed to spot targets in reconnaissance photos which will be tested next year in Defender-Europe 20, and plans to gather more data for AI by equipping RQ-7Bv2 Shadow drones with sensor suites and fielding 200,000 IVAS soldier goggles.

Hurd and Kelly announce new AI initiative with Bipartisan Policy Center: In partnership with the Bipartisan Policy Center, Reps. Hurd and Kelly will develop a national AI strategy aimed at guiding Congress and the executive branch. They plan to convene public and private sector experts to weigh in on the challenges and opportunities of crafting policy on artificial intelligence, concluding with a federal AI framework. Hurd and Kelly previously co-authored a white paper on the importance of AI after hosting a series of congressional hearings.

What We’re Reading

Report: Opinion of the Data Ethics Commission, The Data Ethics Commission of the Federal Government of Germany (October 2019)

Book: The Impact of Emerging Technologies on the Law of Armed Conflict, edited by Eric Talbot Jensen and Major Ronald T. P. Alcala (September 2019)

Post: Artificial Intelligence Research Needs Responsible Publication Norms, Rebecca Crootof in Lawfare (October 2019)

In Translation
CSET's translations of significant foreign language documents on AI

Russia’s National AI Strategy: Decree of the President of the Russian Federation on the Development of Artificial Intelligence in the Russian Federation: Russia’s national strategy for the development of artificial intelligence, released in October 2019. The document sets out a number of short-term (to be completed by 2024) and medium-term (by 2030) qualitative goals designed to build Russia into a leading AI power. For more, see analysis by CSET’s Margarita Konaev.  

What’s New at CSET

TESTIMONYPUBLICATIONSIN THE NEWS
Events


What else is going on?
Suggest stories, documents to translate & upcoming events here.
Blacklisting Chinese AI startups, deepfake detection and NSF funding for AI research
Wednesday, October 16, 2019
Worth Knowing

Trump Administration blacklists top Chinese AI startups for human rights violations: The White House added 28 organizations to the Entity List for human rights abuses in Xinjiang, prohibiting them from buying U.S.-developed technologies. The list named eight tech companies, including Megvii, iFlytek and SenseTime (three of China’s top AI companies), along with Dahua and Hikvision (leading manufacturers of surveillance products).
OpenAI’s AI-powered robot learns to solve a Rubik’s Cube one-handed: OpenAI announced yesterday that it used reinforcement learning to enable a dextrous robot to complete a Rubik’s Cube. While robots have mastered Rubik’s Cubes before, “Dactyl” is unique in that it was trained in a variety of simulated environments and learned to adapt to the real world. The result is a high level of adaptability for a robotic hand, which the researchers believe may be an important step toward general-purpose robots.
Putin signs Russia’s AI strategy: President Vladimir Putin has approved Russia’s national AI strategy through 2030. The strategy, not yet translated into English, centralizes national AI efforts under the Ministry of Economic Development, according to Kommersant. Notably, it defines artificial intelligence as technology that can “simulate human cognitive functions,” including self-learning, and can achieve results comparable to those of “human intellectual activity.”
Google creates a dataset of 3,000 deepfakes to assist with detection efforts: Alphabet’s Jigsaw and Google produced and publicly released the Deep Fake Detection Dataset, with the goal of supporting research efforts to develop automatic deepfake detection. The dataset consists of 1,000 videos altered by four deepfake methods, as well as models to generate new deepfake data. Google’s move comes after Facebook, Microsoft, MIT and others dedicated $10 million to a new deepfake detection dataset and challenge.
Government Updates

NSF creates new program to fund $200M in long-term AI research over six years: The National Science Foundation announced the creation of a new National AI Research Institutes program to advance large-scale, long-term AI research. The grant will fund the creation of research institutes at the nexus of government, industry and academia focused on a series of core priorities in AI. In the first year, NSF anticipates disbursing $120 million in grants. Grant proposals are due to NSF by late January 2020.

DOE announces $13M in new funding for AI research projects: The Department of Energy’s Office of Science announced $13 million in funding for five research projects aimed at “improving AI as a tool of scientific investigation and prediction.” The projects fund both universities and DOE national laboratories. This development builds on other recent AI-related activities at the DOE, including the creation of the Artificial Intelligence and Technology Office and funding from ARPA-E for AI-developed tools to reduce nuclear power plant operating expenses.

Keep STEM Talent Act of 2019 introduced in the House: On October 4th, Reps. Bill Foster and Eddie Bernice Johnson introduced the Keep STEM Talent Act of 2019, which would provide permanent-resident status to advanced STEM degree holders. The bill, a companion to one introduced in the Senate by Sen. Richard Durbin, exempts STEM graduates from annual green card caps. Long wait times for green cards were identified as a key problem facing international AI talent in CSET’s recent report on Immigration Policy and the U.S. AI Sector.

What We’re Reading

Article: Chinese Semiconductor Industrial Policy: Past and Present and Prospects for Future Success, John VerWey in the United States International Trade Commission (July and August 2019), and The South Korea-Japan Trade Dispute in Context: Semiconductor Manufacturing, Chemicals, and Concentrated Supply Chains by Samuel M. Goodman, Dan Kim and John VerWey

Book: The Transpacific Experiment: How China and California Collaborate and Compete for Our Future, Matt Sheehan (August 2019)

Post: Critiquing Carnegie’s AI Surveillance Paper, John Honovick and Charles Rollet at IPVM (September 2019) — a response to a Carnegie paper we included in the previous edition of policy.ai

In Translation
CSET's translations of significant foreign language documents on AI

Municipal action plan for “military-civil fusion” in developing “intelligent technologies”: Tianjin Municipal Action Plan for Military-Civil Fusion Special Projects in Intelligent Technology. While many Chinese local governments have published “military-civil fusion” plans, Tianjin’s is among the most detailed. It provides insight into local efforts to steer the development of emerging technologies in directions that fulfill PLA requirements.

What’s New at CSET

JOB OPENINGS
CSET is hiring! We’re looking to fill the following roles:
  • Data Scientist: Apply machine learning, NLP, predictive modeling, and other advanced computational techniques to evaluate hypotheses and derive data-driven insights. BA in relevant area and 6+ yrs of experience required.
  • Research Project Manager: Coordinate hiring of student research assistants; manage a research pipeline. BA in relevant area and 2+ yrs project management experience required.
  • Research Analyst: Collaborate with Research Fellows to develop and execute research projects. 2+ yrs and BA in relevant area required (MA preferred).
PUBLICATIONSIN THE NEWS
Events


What else is going on?
Suggest stories, documents to translate & upcoming events here.
OpenAI experiments with hide-and-seek, California bans facial recognition use by police, and members of Congress seek to revive the OTA
Wednesday, October 2, 2019
Worth Knowing

OpenAI experiment tests how AI might “evolve” through competition: Researchers observed teams of AI agents playing billions of games of hide-and-seek in an attempt to understand emergent behavior. Over time, agents learned to use available tools in increasingly complex ways — including adopting strategies that programmers did not expect. The researchers hope this type of reinforcement learning will allow AI systems to solve increasingly complex problems in the future, but found the number of repetitions required makes it difficult to apply this technique to real-world settings.
California legislature bans facial recognition use by law enforcement: Both houses of the California legislature passed AB-1215, which prohibits law enforcement from “installing, activating, or using any biometric surveillance system in connection with an officer camera” for three years. The bill now goes to Gov. Gavin Newsom; if he signs it, California will become the largest state to ban specific uses of facial recognition.
Kalaris Conference convenes AI and national security experts: U.S. security interests will suffer if the United States doesn’t work with its allies to invest wisely in AI capabilities, leading figures from the intelligence and defense communities said at the Kalaris Intelligence Conference last week. CSET and Georgetown University’s Center for Security Studies co-hosted the annual conference. Among the speakers were Sue Gordon, former Principal Deputy Director of National Intelligence, and Lt. Gen. Jack Shanahan, director of the DoD’s Joint Artificial Intelligence Center.
France releases military AI strategy: The French Ministry of the Armies has published a comprehensive report on military use of AI, building on the strategy laid out by Minister of Armies Florence Parly in April. The document establishes a Coordination Unit for Defense AI and a ministerial AI ethics committee, and it commits 430 million euros ($470 million) to AI research by 2025. The strategy describes the United States and China as global leaders in AI, but outlines a possible role for France if it coordinates with other nations within and outside the EU.
Government Updates

JAIC and GSA announce partnership to expand Pentagon’s use of AI: The General Services Administration and the DoD’s Joint AI Center announced their new partnership through the GSA’s Center of Excellence initiative on September 25th. They aim to expand the Pentagon’s use of AI by accelerating the delivery of AI capabilities and modernizing programmatic and acquisition processes. GSA also hopes the partnership will spur greater AI adoption across government.

Office of Technology Assessment Improvement and Enhancement Act introduced in House and Senate: Reps. Takano and Foster and Sens. Tillis and Hirono introduced the bipartisan Office of Technology Assessment Improvement and Enhancement Act in both houses of Congress on September 19th. It would revamp, rename and improve the OTA — which was defunded in 1995 — to ensure nonpartisan technology assessment is available to Congress. The House FY20 Legislative Branch Appropriations bill includes $6 million to fund the OTA; the Senate bill defers the issue until an ongoing report is complete.

MQ-25, the Navy’s unmanned refueling drone, completes successful test flight: The U.S. Navy announced a successful first test flight for its autonomous refueling drone, the Boeing-developed MQ-25 Stingray. The two-hour flight included autonomous taxi and takeoff. The Navy plans to integrate the MQ-25 into its strike arm by 2024 as the first carrier-launched autonomous unmanned aircraft.

Additional $8 million for NIST AI research included in Senate appropriations bill: The Senate FY20 Commerce, Justice and Science Appropriations bill includes $1.04 billion for the National Institute of Standards and Technology, a $52.5 million increase above the FY19 enacted level. Of the additional funds, $8 million would be allocated to expand NIST’s AI research and measurement efforts, including developing resources to model AI behavior and train, test and compare AI systems.

In Translation
CSET's translations of significant foreign language documents on AI

The emerging technologies of interest to China’s military: Guidelines for Basic Research and Cutting-Edge Technology Projects (2018): Issued by China’s State Administration of Science, Technology and Industry for National Defense, this document identifies several emerging technologies of interest to the Chinese military. SASTIND circulated these guidelines to Chinese universities and research institutes in 2018 to encourage them to apply for grants to conduct basic research in PLA areas of interest.

What’s New at CSET

IN THE NEWS
Events


What else is going on?
Suggest stories, documents to translate & upcoming events here.
AI professors leave for industry, Russia drafts AI strategy, and the White House requests $1B for AI R&D
Wednesday, September 18, 2019
Worth Knowing

Study finds flow of AI professors to industry discourages innovation: New research from the University of Rochester shows that after AI professors leave academia for private-sector work, fewer of their students start AI companies. The study, first covered by The New York Times, found that about 10 times as many North American professors left for tech companies in 2018 as did in 2009. The researchers say this trend could eventually hamper AI innovation and the economy.
Preview of Russia’s AI strategy: President Vladimir Putin is reviewing a draft AI strategy that he ordered state-owned Sberbank to prepare, according to DefenseOne. The wide-ranging document covers fundamental investments in AI — including funding for research, ethical and data regulations, and hardware and software developments — as well as specific applications of AI in healthcare and education. The final version is expected next month.
Record number of submitted papers for NeurIPS: The Conference on Neural Information Processing Systems received 6,743 paper submissions this year — up 39% from last year, and double the number of submissions in 2017. Google, MIT and Stanford were the most common institutional affiliations across accepted papers. Since NeurIPS is the largest annual AI conference, its metrics are often used as indicators of continued growth and enthusiasm in the field of machine learning.
Partnership on AI calls for improving immigration policies that affect AI experts: The Partnership on AI, which connects major tech companies with government and NGOs to create collaborative proposals, released a report last week calling for increasing access to visas and other immigration benefits for global AI/ML experts. The report lays out policy recommendations including streamlined visa reviews for highly skilled individuals, AI/ML visa classifications and visa categories for AI/ML students and interns.
Government Updates

White House requests $1B in non-defense AI spending in 2020: The Trump administration submitted a supplemental request for $973.5 million in non-defense AI R&D spending for fiscal year 2020. While this number is higher than previous years, some industry leaders say it’s not enough. Looking ahead, a White House memo listed AI as a priority for the 2021 R&D budget.

Senate Defense Appropriations bill boosts defense funding for AI: The Senate Appropriations Committee approved its 2020 Defense Appropriations bill, which now awaits Senate consideration. The bill supports the President’s budget request for the Joint Artificial Intelligence Center at $208.8 million. In addition, it provides $83.5 million above the President’s budget on accounts labeled for AI-related Research, Development, Test and Evaluation.

White House holds AI Summit: On September 9th, the White House held The Summit on AI in Government for 175 industry, government and academic experts in AI. The event concluded with three case studies of AI use to improve government operations.

Air Force releases 2019 AI Strategy: On September 12th, the Air Force released its Annex to the DoD AI Strategy issued in 2018. The Annex aligns USAF strategy with that of DoD and focuses on expanding access to AI, preparing an AI workforce and treating data as a strategic asset.

What We’re Reading

Attacking Artificial Intelligence: AI’s Security Vulnerability and What Policymakers Can Do About It, Belfer Center for Science and International Affairs, Harvard Kennedy School (August 2019)

A Tentative Framework for Examining U.S. and Chinese Expenditures for Research and Development on Artificial Intelligence, The Institute for Defense Analyses Science & Technology Policy Institute (September 2019)

In Translation
CSET's translations of significant foreign language documents on AI

Open-Source AI development platforms: Guidance on National New Generation Artificial Intelligence Open Innovation Platform Construction Work: Chinese Ministry of Science and Technology document describing the updated approval process for Chinese AI tech companies’ “open innovation platforms.” This document builds on the 2017 AI Development Plan, which identified open-source platforms as crucial to making China the world leader in AI by 2030.

What’s New at CSET

REPORTSIN THE NEWS
Events


What else is going on?
Suggest stories, documents to translate & upcoming events here.