Monthly Archives: March 2024

A former Google engineer was charged with stealing AI technology while secretly working with two China-based companies. Carl Court/Getty Images hide caption

A former Google engineer was charged with stealing AI technology while secretly working with two China-based companies.

A Chinese national who allegedly stole more than 500 files from Google with confidential information on the company's AI technology has been arrested and charged with stealing trade secrets, according to the Justice Department.

The defendant, former Google employee Linwei Ding, was arrested Wednesday morning in Newark, Calif. The 38-year-old faces four counts of theft of trade secrets. Prosecutors say at the same time that Ding was working for Google and stealing the building blocks of its AI technology, he was also secretly employed by two China-based tech companies.

"The Justice Department will not tolerate the theft of artificial intelligence and other advanced technologies that could put our national security at risk," Attorney General Merrick Garland said in a statement. "We will fiercely protect sensitive technologies developed in America from falling into the hands of those who should not have them."

The case is latest example of what American officials say is a relentless campaign by China to try to steal U.S. trade secrets, technology and intellectual property. Officials say China aims to use those stolen secrets to supplant the U.S. as the world's leading power.

"Today's charges are the latest illustration of the lengths affiliates of companies based in the People's Republic of China are willing to go to steal American innovation," said FBI Director Christopher Wray. "The theft of innovative technology and trade secrets from American companies can cost jobs and have devastating economic and national security consequences."

The U.S. is the global leader in AI, an emerging technology that could reshape many facets of modern life.

AI also could become an indispensable tool to help law enforcement protect public safety. But Justice Department officials also have warned of the potential dangers that AI poses to national security if it falls into the hands of criminals or hostile nation states.

The department has also formed a unit to protect advanced American technology such as AI from being pilfered by foreign adversaries.

In Ding's case, the indictment says the trade secrets he allegedly stole are related to "the hardware infrastructure and software platform that allow Google's supercomputing data centers to train large AI models through machine learning."

Google spokesperson Jose Castaneda said the company has "strict safeguards to prevent theft of our confidential commercial information and trade secrets."

"After an investigation, we found that this employee stole numerous documents, and we quickly referred the case to law enforcement," Castaneda said. "We are grateful to the FBI for helping protect our information and will continue cooperating with them closely."

The indictment says Ding was hired at Google as a software engineer in 2019. His work focused on the development of software related to machine learning and AI applications, according to prosecutors.

In May of 2022, Ding allegedly began uploading confidential informationmore than 500 unique files in allfrom Google's network into a personal Google Cloud account.

Prosecutors say Ding tried to hide what he was doing by copying the stolen files first into the Apple Notes application on his laptop, converting them into PDF files and uploading those into his personal Cloud account.

Less than a month later, court papers say, Ding received emails from the head of a Chinese technology company, Beijing Rongshu Lianzhi Technology, with an offer to be the company's chief technology officer.

Ding allegedly traveled to China to help raise money for the company, which worked on AI, and was announced as the company's CTO. A year later, Ding also allegedly founded his own technology company, Zhisuan, that also focused on AI and machine learning.

Prosecutors say Ding never informed Google of his ties to either Chinese company, and continued to be employed by Google.

Then in December 2023, court papers say, Google detected Ding trying to upload more files from the company's network to his personal account while he was in China. Ding allegedly told the company's investigator that he'd uploaded the files as evidence of his work for Google.

A week after being interviewed by the investigator, Ding allegedly booked a one-way ticket to Beijing. He then sent his resignation letter to Google. Shortly after that, the company learned of Ding's role with Zhisuan. Google then suspended his access to the company's networks.

Shortly after that, the FBI began its investigation.

See more here:

Chinese national arrested and charged with stealing AI trade secrets from Google - NPR

By Chris Dally, Business Designation Owner AWS By Victor Rojo, Technical Designation Lead AWS By Chris Butler, Sr. Product Manager, Launch AWS By Justin Freeman, Sr. Partner Development Specialist, Catalyst AWS

In todays rapidly evolving technology landscape, generative artificial intelligence (AI) is leading the charge in innovation, revolutionizing the way organizations work. According to a McKinsey report, generative AI could account for over 75% of total yearly AI value, with high expectations for major or disruptive change in industries. Additionally, the report states generative AI technologies have the potential to automate work activities that absorb 60-70% of employees time.

With the ability to automate tasks, enhance productivity, and enable hyper-personalized customer experiences, businesses are seeking specialized expertise to build a successful generative AI strategy.

To support this need, were excited to announce the AWS Generative AI Competencyan AWS Specialization that helps Amazon Web Services (AWS) customers more quickly adopt generative AI solutions and strategically position themselves for the future. AWS Generative AI Competency Partners provide a full range of services, tools, and infrastructurewith tailored solutions in areas like security, applications, and integrations to give customers flexibility and choice across models and technologies.

Partners play an important role in supporting AWS customers leveraging our comprehensive suite of generative AI services. We are excited to recognize and highlight partners with proven customer success with generative AI on AWS through the AWS Generative AI Competency, making it easier for our customers to find and identify the right partners to support their unique needs. ~ Swami Sivasubramanian, Vice President of Database, Analytics and ML, AWS

According to Canalys, AWS is the first to launch a Generative AI competency for partners. By validating the partners business and technical expertise in this way, AWS customers are able to invest with greater confidence in generative AI solutions from these partners. This new competency is a critical entry point into the Generative AI partner opportunity, which Canalys estimates will grow to US$158 billion by 2028.

Generative AI has truly ushered in a new era of innovation and transformative value across both business and technology. A recent Canalys study found that 87% of customers rank partner specializations as a top three selection criteria. With the AWS Generative AI Competency launch, were helping customers take advantage of the capabilities that our technically validated Generative AI Partners have to offer. ~ Ruba Borno, Vice President of AWS Worldwide Channels and Alliances

Leveraging AI technologies such as Amazon Bedrock, Amazon SageMaker JumpStart, AWS Trainium, AWS Inferentia, and accelerated computing instances on Amazon Elastic Compute Cloud (Amazon EC2), AWS Generative AI Competency Partners have deep expertise building and deploying groundbreaking applications across industries, including healthcare and life sciences, media and entertainment, public sector, and financial services.

We invite you to explore the following AWS Generative AI Competency Launch Partner offerings recommended by AWS.

These AWS Partners have deep expertise working with businesses to help them adopt and strategize generative AI, build and test generative AI applications, train and customize foundation models, operate, support, and maintain generative AI applications and models, protect generative AI workloads, and define responsible AI principles and frameworks.

These AWS Partners utilize foundation models (FMs) and related technologies to automate domain-specific functions, enhancing customer differentiation across all business lines and operations. Partners fall into three categories: Generative AI applications, Foundation Models and FM-based Application Development, and Infrastructure and Data.

AWS Generative AI Competency Partners make it easier for customers to innovate with enterprise-grade security and privacy, foundation models, generative AI-powered applications, a data-first approach, and a high-performance, low-cost infrastructure.

Explore the AWS Generative AI Partners page to learn more.

AWS Partners with Generative AI offerings can learn more about becoming an AWS Competency Partner.

AWS Specialization Partners gain access to strategic and confidential content, including product roadmaps, feature release previews, and demos, as part of the AWS PartnerEquip event series. To attend live events in your region or tune in virtually, register for an upcoming session. In addition to AWS Specialization Program benefits, AWS Generative AI Competency Partners receive unique benefits such as bi-annual strategy sessions to aid joint sales motions. To learn more, review the AWS Specialization Program Benefits Guide in AWS Partner Central (login required).

AWS Partners looking to get their Generative AI offering validated through the AWS Competency Program must be validated or differentiated members of the Software or Services Path prior to applying.

To apply, please review the Program Guide and access the application in AWS Partner Central.

Follow this link:

Revolutionize Your Business with AWS Generative AI Competency Partners | Amazon Web Services - AWS Blog

Key Takeaways

Micron Technology(MU) shares rose to a record high Thursday as analysts from Goldman Sachs and Stifel raised their price targets on the stock, citing the company's position amid the artificial intelligence (AI) boom.

Shares of Micron closed 3.6% higher at $98.98 Thursday, contributing to a more than 20% increase since the start of 2024.

Goldman Sachs analysts raised their price target for Micron to $112 from $103 with a "buy" rating, saying that the company is an "under-appreciated AI beneficiary."

"We believe Micron is well-positioned to benefit from the proliferation of AI across data centers (i.e. the core) and the edge (e.g. PCs, smartphones) as demand for more compute drives an increase in content," they said.

The analysts noted that the stock's year-to-date gains were more muted compared to those of some of its peers in the compute and networking space, nodding to Nvidia (NVDA) and Arm (ARM). Nvidia shares have nearly doubled while Arm shares have more than doubled in value since the start of 2024.

Stifel analysts indicated that the firm believes consensus estimates are "wrong and too low," adding that it anticipates Micron "breaking out to higher highs, perhaps aided by the most compelling growth-valuation ratio amongst larger cap 'AI' relevant stocks."

The analysts upgraded the stock to a "buy" rating from "hold" and increased its price targetto $120 from $80.

Stifel said that Micron's position amid the AI boom drove the stock upgrade. Generative AI (GenAI) needs high bandwidth memory (HBM), "and Micron now has a seat at the table," Stifel analysts wrote.

Micron announced in February that it began mass production of an HBM chip for Nvidias AIgraphic processing units (GPUs), bolstering its position in the AI ecosystem.

See the rest here:

Micron Hits Record High With Analysts Calling It an 'Under-Appreciated AI Beneficiary' - Investopedia

New York City has quietly filled the role of director of artificial intelligence and machine learning, City & State has learned. In mid-January, Jiahao Chen, a former director of AI research at JPMorgan Chase and the founder of independent consulting company Responsible AI LLC, took on the role, which has been described by the citys Office of Technology and Innovation as spearheading the citys comprehensive AI strategy.

Despite Mayor Eric Adams administration publicizing the position last January, Chens hiring nearly a year later came without any fanfare or even an announcement. The first mention of Chen as director of AI came in a press release sent out by the Office of Technology and Innovation on Thursday morning, announcing next steps in the citys AI Action Plan. OTI Director of AI and Machine Learning Jiahao Chen will manage implementation of the Action Plan, the press release noted.

New York City previously had an AI director under former Mayor Bill de Blasios administration. Neal Parikh served as the citys director of AI under the office of former Chief Technology Officer John Paul Farmer, which released a citywide AI strategy in 2021. Under de Blasio, the city also had an algorithms management and policy officer to guide the city in the development, responsible use and assessment of algorithmic tools, which can include AI and machine learning. The old CTOs office and the work of the algorithms officer was consolidated along with the citys other technology-related offices into the new Office of Technology and Innovation at the outset of the Adams administration.

The Adams administration has referred to its own director of AI and machine learning as a new role, however, and has suggested that the position will be more empowered, in part because it is under the larger, centralized Office of Technology and Innovation. According to the job posting last January, which noted a $75,000 to $140,000 pay range, the director will be responsible for helping agencies use AI and machine learning tools responsibly, consulting with agencies on questions about AI use and governance, and serving as a subject matter expert on citywide policy and planning, among other things. How the role will actually work in practice remains to be seen.

The Adams administrations AI action plan was published in October, and isa 37-point road map aimed at helping the city responsibly harness the power of AI for good. On Thursday, the Office of Technology and Innovation announced the first update on the action plan, naming members of an advisory network that will consult on the citys work. That list includes former City Council Member Marjorie Velzquez, who is now vice president of policy at Tech:NYC. The office also released a set of AI principles and definitions, and guidance on generative AI.

OTI spokesperson Ray Legendre said that an offer for the position of director of AI was extended to Chen before the citys hiring freeze began last October. The office did not explicitly address why Chens hiring wasnt announced when he started the role. Over the past two months, Jiahao has been a key part of our ongoing efforts to implement the AI Action Plan, Legendre wrote in an email. Our focus at OTI over the past few months has been on making progress on the Action Plan which is what we announced today.

According to the website for Responsible AI LLC, Chens independent consulting company, Chens resume includes stints in academia as well as the private sector, including as a senior manager of data science at Capital One, and as director of AI research at JPMorgan Chase.

After City & State inquired about Chens role, Chen confirmed it on X, writing I can finally talk about my new job!

Visit link:

The Adams administration quietly hired its first AI czar. Who is he? - City & State New York

Crabtree, G. Self-driving laboratories coming of age. Joule 4, 25382541 (2020).

Article CAS Google Scholar

Wang, H. et al. Scientific discovery in the age of artificial intelligence. Nature 620, 4760 (2023). This review explores how AI can be incorporated across the research pipeline, drawing from a wide range of scientific disciplines.

Article CAS PubMed Google Scholar

Dillion, D., Tandon, N., Gu, Y. & Gray, K. Can AI language models replace human participants? Trends Cogn. Sci. 27, 597600 (2023).

Article PubMed Google Scholar

Grossmann, I. et al. AI and the transformation of social science research. Science 380, 11081109 (2023). This forward-looking article proposes a variety of ways to incorporate generative AI into social-sciences research.

Article CAS PubMed Google Scholar

Gil, Y. Will AI write scientific papers in the future? AI Mag. 42, 315 (2022).

Google Scholar

Kitano, H. Nobel Turing Challenge: creating the engine for scientific discovery. npj Syst. Biol. Appl. 7, 29 (2021).

Article PubMed PubMed Central Google Scholar

Benjamin, R. Race After Technology: Abolitionist Tools for the New Jim Code (Oxford Univ. Press, 2020). This book examines how social norms about race become embedded in technologies, even those that are focused on providing good societal outcomes.

Broussard, M. More Than a Glitch: Confronting Race, Gender, and Ability Bias in Tech (MIT Press, 2023).

Noble, S. U. Algorithms of Oppression: How Search Engines Reinforce Racism (New York Univ. Press, 2018).

Bender, E. M., Gebru, T., McMillan-Major, A. & Shmitchell, S. On the dangers of stochastic parrots: can language models be too big? in Proc. 2021 ACM Conference on Fairness, Accountability, and Transparency 610623 (Association for Computing Machinery, 2021). One of the first comprehensive critiques of large language models, this article draws attention to a host of issues that ought to be considered before taking up such tools.

Crawford, K. Atlas of AI: Power, Politics, and the Planetary Costs of Artificial Intelligence (Yale Univ. Press, 2021).

Johnson, D. G. & Verdicchio, M. Reframing AI discourse. Minds Mach. 27, 575590 (2017).

Article Google Scholar

Atanasoski, N. & Vora, K. Surrogate Humanity: Race, Robots, and the Politics of Technological Futures (Duke Univ. Press, 2019).

Mitchell, M. & Krakauer, D. C. The debate over understanding in AIs large language models. Proc. Natl Acad. Sci. USA 120, e2215907120 (2023).

Article PubMed PubMed Central Google Scholar

Kidd, C. & Birhane, A. How AI can distort human beliefs. Science 380, 12221223 (2023).

Article CAS PubMed Google Scholar

Birhane, A., Kasirzadeh, A., Leslie, D. & Wachter, S. Science in the age of large language models. Nat. Rev. Phys. 5, 277280 (2023).

Article Google Scholar

Kapoor, S. & Narayanan, A. Leakage and the reproducibility crisis in machine-learning-based science. Patterns 4, 100804 (2023).

Article PubMed PubMed Central Google Scholar

Hullman, J., Kapoor, S., Nanayakkara, P., Gelman, A. & Narayanan, A. The worst of both worlds: a comparative analysis of errors in learning from data in psychology and machine learning. In Proc. 2022 AAAI/ACM Conference on AI, Ethics, and Society (eds Conitzer, V. et al.) 335348 (Association for Computing Machinery, 2022).

Rudin, C. Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead. Nat. Mach. Intell. 1, 206215 (2019). This paper articulates the problems with attempting to explain AI systems that lack interpretability, and advocates for building interpretable models instead.

Article PubMed PubMed Central Google Scholar

Crockett, M. J., Bai, X., Kapoor, S., Messeri, L. & Narayanan, A. The limitations of machine learning models for predicting scientific replicability. Proc. Natl Acad. Sci. USA 120, e2307596120 (2023).

Article CAS PubMed PubMed Central Google Scholar

Lazar, S. & Nelson, A. AI safety on whose terms? Science 381, 138 (2023).

Article PubMed Google Scholar

Collingridge, D. The Social Control of Technology (St Martins Press, 1980).

Wagner, G., Lukyanenko, R. & Par, G. Artificial intelligence and the conduct of literature reviews. J. Inf. Technol. 37, 209226 (2022).

Article Google Scholar

Hutson, M. Artificial-intelligence tools aim to tame the coronavirus literature. Nature https://doi.org/10.1038/d41586-020-01733-7 (2020).

Article PubMed Google Scholar

Haas, Q. et al. Utilizing artificial intelligence to manage COVID-19 scientific evidence torrent with Risklick AI: a critical tool for pharmacology and therapy development. Pharmacology 106, 244253 (2021).

Article CAS PubMed Google Scholar

Mller, H., Pachnanda, S., Pahl, F. & Rosenqvist, C. The application of artificial intelligence on different types of literature reviews a comparative study. In 2022 International Conference on Applied Artificial Intelligence (ICAPAI) https://doi.org/10.1109/ICAPAI55158.2022.9801564 (Institute of Electrical and Electronics Engineers, 2022).

van Dinter, R., Tekinerdogan, B. & Catal, C. Automation of systematic literature reviews: a systematic literature review. Inf. Softw. Technol. 136, 106589 (2021).

Article Google Scholar

Aydn, . & Karaarslan, E. OpenAI ChatGPT generated literature review: digital twin in healthcare. In Emerging Computer Technologies 2 (ed. Aydn, .) 2231 (zmir Akademi Dernegi, 2022).

AlQuraishi, M. AlphaFold at CASP13. Bioinformatics 35, 48624865 (2019).

Article CAS PubMed PubMed Central Google Scholar

Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583589 (2021).

Article CAS PubMed PubMed Central Google Scholar

Lee, J. S., Kim, J. & Kim, P. M. Score-based generative modeling for de novo protein design. Nat. Computat. Sci. 3, 382392 (2023).

Article CAS Google Scholar

Gmez-Bombarelli, R. et al. Design of efficient molecular organic light-emitting diodes by a high-throughput virtual screening and experimental approach. Nat. Mater. 15, 11201127 (2016).

Article PubMed Google Scholar

Krenn, M. et al. On scientific understanding with artificial intelligence. Nat. Rev. Phys. 4, 761769 (2022).

Article PubMed PubMed Central Google Scholar

Extance, A. How AI technology can tame the scientific literature. Nature 561, 273274 (2018).

Article CAS PubMed Google Scholar

Hastings, J. AI for Scientific Discovery (CRC Press, 2023). This book reviews current and future incorporation of AI into the scientific research pipeline.

Ahmed, A. et al. The future of academic publishing. Nat. Hum. Behav. 7, 10211026 (2023).

Article PubMed Google Scholar

Gray, K., Yam, K. C., ZhenAn, A. E., Wilbanks, D. & Waytz, A. The psychology of robots and artificial intelligence. In The Handbook of Social Psychology (eds Gilbert, D. et al.) (in the press).

Argyle, L. P. et al. Out of one, many: using language models to simulate human samples. Polit. Anal. 31, 337351 (2023).

Article Google Scholar

Aher, G., Arriaga, R. I. & Kalai, A. T. Using large language models to simulate multiple humans and replicate human subject studies. In Proc. 40th International Conference on Machine Learning (eds Krause, A. et al.) 337371 (JMLR.org, 2023).

Binz, M. & Schulz, E. Using cognitive psychology to understand GPT-3. Proc. Natl Acad. Sci. USA 120, e2218523120 (2023).

Article CAS PubMed PubMed Central Google Scholar

Ornstein, J. T., Blasingame, E. N. & Truscott, J. S. How to train your stochastic parrot: large language models for political texts. Github, https://joeornstein.github.io/publications/ornstein-blasingame-truscott.pdf (2023).

He, S. et al. Learning to predict the cosmological structure formation. Proc. Natl Acad. Sci. USA 116, 1382513832 (2019).

Article MathSciNet CAS PubMed PubMed Central Google Scholar

Mahmood, F. et al. Deep adversarial training for multi-organ nuclei segmentation in histopathology images. IEEE Trans. Med. Imaging 39, 32573267 (2020).

Article PubMed PubMed Central Google Scholar

Teixeira, B. et al. Generating synthetic X-ray images of a person from the surface geometry. In Proc. IEEE Conference on Computer Vision and Pattern Recognition 90599067 (Institute of Electrical and Electronics Engineers, 2018).

Marouf, M. et al. Realistic in silico generation and augmentation of single-cell RNA-seq data using generative adversarial networks. Nat. Commun. 11, 166 (2020).

Article CAS PubMed PubMed Central Google Scholar

Watts, D. J. A twenty-first century science. Nature 445, 489 (2007).

Article CAS PubMed Google Scholar

boyd, d. & Crawford, K. Critical questions for big data. Inf. Commun. Soc. 15, 662679 (2012). This article assesses the ethical and epistemic implications of scientific and societal moves towards big data and provides a parallel case study for thinking about the risks of artificial intelligence.

Article Google Scholar

Jolly, E. & Chang, L. J. The Flatland fallacy: moving beyond lowdimensional thinking. Top. Cogn. Sci. 11, 433454 (2019).

Article PubMed Google Scholar

Yarkoni, T. & Westfall, J. Choosing prediction over explanation in psychology: lessons from machine learning. Perspect. Psychol. Sci. 12, 11001122 (2017).

Article PubMed PubMed Central Google Scholar

Radivojac, P. et al. A large-scale evaluation of computational protein function prediction. Nat. Methods 10, 221227 (2013).

Article CAS PubMed PubMed Central Google Scholar

Bileschi, M. L. et al. Using deep learning to annotate the protein universe. Nat. Biotechnol. 40, 932937 (2022).

Article CAS PubMed Google Scholar

Barkas, N. et al. Joint analysis of heterogeneous single-cell RNA-seq dataset collections. Nat. Methods 16, 695698 (2019).

Article CAS PubMed PubMed Central Google Scholar

Demszky, D. et al. Using large language models in psychology. Nat. Rev. Psychol. 2, 688701 (2023).

Article Google Scholar

Karjus, A. Machine-assisted mixed methods: augmenting humanities and social sciences with artificial intelligence. Preprint at https://arxiv.org/abs/2309.14379 (2023).

Davies, A. et al. Advancing mathematics by guiding human intuition with AI. Nature 600, 7074 (2021).

Article CAS PubMed PubMed Central Google Scholar

Peterson, J. C., Bourgin, D. D., Agrawal, M., Reichman, D. & Griffiths, T. L. Using large-scale experiments and machine learning to discover theories of human decision-making. Science 372, 12091214 (2021).

Article CAS PubMed Google Scholar

Ilyas, A. et al. Adversarial examples are not bugs, they are features. Preprint at https://doi.org/10.48550/arXiv.1905.02175 (2019)

Semel, B. M. Listening like a computer: attentional tensions and mechanized care in psychiatric digital phenotyping. Sci. Technol. Hum. Values 47, 266290 (2022).

Article Google Scholar

Gil, Y. Thoughtful artificial intelligence: forging a new partnership for data science and scientific discovery. Data Sci. 1, 119129 (2017).

Read this article:

Artificial intelligence and illusions of understanding in scientific research - Nature.com