Microsoft's Backdoor Research, Israel-Azerbaijan AI Pact, and EU AI Act Consortium - AI Policy Brief #55
February 10, 2026 -
AI Policy Brief
Hi there,
Welcome to this week's edition of the AI Policy Brief, your go-to source for the latest developments in AI regulations, safety standards, and compliance requirements worldwide. This week, we delve into a variety of significant updates, starting with Microsoft's new scanner designed to detect backdoors in language models, underscoring the growing need for integrity protections in AI systems. Meanwhile, Pennsylvania Governor Josh Shapiro has introduced a comprehensive plan to manage the state's data center expansion, focusing on sustainability and community impact.
In other news, DeepKeep has launched its AI security platform on major cloud marketplaces like Google Cloud and AWS, aiming to bolster AI safety and compliance for businesses. Additionally, Ghana has announced tech-law reforms to support AI and fintech growth, while Israel and Azerbaijan have signed a memorandum of understanding to enhance AI collaboration. These stories, along with many others, highlight the dynamic landscape of AI policy and regulation across the globe.
Welcome to this week's edition of the AI Policy Brief, your go-to source for the latest developments in AI regulations, safety standards, and compliance requirements worldwide. This week, we delve into a variety of significant updates, starting with Microsoft's new scanner designed to detect backdoors in language models, underscoring the growing need for integrity protections in AI systems. Meanwhile, Pennsylvania Governor Josh Shapiro has introduced a comprehensive plan to manage the state's data center expansion, focusing on sustainability and community impact.
In other news, DeepKeep has launched its AI security platform on major cloud marketplaces like Google Cloud and AWS, aiming to bolster AI safety and compliance for businesses. Additionally, Ghana has announced tech-law reforms to support AI and fintech growth, while Israel and Azerbaijan have signed a memorandum of understanding to enhance AI collaboration. These stories, along with many others, highlight the dynamic landscape of AI policy and regulation across the globe.
Microsoft Research on Backdoored Language Models
Researchers from Microsoft have published findings on detecting backdoors in language models, introducing a scanner for identifying compromised models efficiently. Read more
Governor Shapiro Introduces Data Center Management Plan
Pennsylvania Governor Josh Shapiro is introducing new guidelines to manage the growth of data centers in the state, focusing on power, community, hiring, and environmental standards. Read more
DeepKeep Launches AI Security Platform on Cloud Marketplaces
DeepKeep has introduced its enterprise AI security platform on Google Cloud and AWS Marketplaces, focusing on enhancing AI safety and compliance for businesses. Read more
Ghana Announces Tech-Law Reforms for AI and Fintech
At the TechLaw Conference 2026, Ghana's Deputy Minister announced plans to enhance legal frameworks to support AI, fintech, and cybersecurity growth. Read more
Israel and Azerbaijan Sign AI Collaboration Agreement
On February 3, 2026, Israel and Azerbaijan signed a memorandum of understanding to enhance collaboration in artificial intelligence, focusing on supercomputing, AI applications, and joint research. Read more
AI Provisions in FY 2026 Spending Package Passed by Congress
The U.S. House of Representatives has passed a spending package that includes provisions for artificial intelligence. The package covers sectors such as defense, financial services, and healthcare, and allocates funding for AI research and cybersecurity. Read more
Irish Government Proposes AI Regulation Bill 2026
The Irish Government has introduced the General Scheme for the Regulation of Artificial Intelligence Bill 2026, aiming to implement the EU AI Act in Ireland. It includes the creation of an independent AI Office and outlines penalties for non-compliance. Read more
Alice Launches Caterpillar Tool for OpenClaw Security
On February 4, 2026, Alice introduced Caterpillar, a free open source tool designed to enhance security for AI agents in OpenClaw by identifying malicious behaviors in skills used by over 6,000 users. Read more
Alabama Lawmakers Introduce AI Regulation Bills
Lawmakers in Alabama have proposed new bills to regulate AI, focusing on health care oversight, content disclosure, and age verification for AI chatbots. Read more
Bipartisan AI Grand Challenges Act Reintroduced
U.S. lawmakers have reintroduced the AI Grand Challenges Act, aiming to use artificial intelligence to address complex issues in sectors like health and cybersecurity. The bill proposes a program for prize competitions to foster innovation. Read more
Varonis Acquires AllTrue.ai for AI Security Management
Varonis has announced its acquisition of AllTrue.ai, aiming to enhance AI security management across enterprises. The acquisition will provide organizations with improved tools for managing and securing AI systems, ensuring compliance and minimizing risks. Read more
Bounteous and Anthropic Launch Claude Code Lab Series
Bounteous has partnered with Anthropic to introduce the Claude Code Lab series, focusing on responsible AI adoption. The series will educate professionals on implementing Claude Code in enterprise settings, with initial events scheduled in Frisco, Texas, and London in 2026. Read more
FAR.AI to Lead EU AI Act CBRN Risk Consortium
The European Commission's AI Office has appointed FAR.AI to head a consortium, including SecureBio, to research technical safety for the EU's AI Act, focusing on CBRN threats. Read more
OpenAI Introduces Trusted Access for Cyber
OpenAI has launched Trusted Access for Cyber, a framework designed to improve AI security by promoting responsible use of advanced AI capabilities. The initiative includes $10 million in API credits to support cybersecurity efforts. Read more
White House Develops AI Security Policy Framework
The White House Office of the National Cyber Director is working on an AI security policy framework to enhance security in U.S.-led AI technologies, as announced by National Cyber Director Sean Cairncross. Read more
University of Warwick Secures AI Research Funding
The University of Warwick has received funding from Responsible AI UK to explore safe and effective AI deployment. This initiative is part of a larger collaboration grant aimed at enhancing AI governance frameworks. Read more
Hong Kong Monetary Authority Unveils Fintech Promotion Blueprint
The Hong Kong Monetary Authority has released a Fintech Promotion Blueprint to encourage responsible innovation in finance, focusing on AI and DLT integration. Read more
DARPA-Funded AI Trust Study at Boston University
An 18-month DARPA-funded project led by Joshua Peterson at Boston University aims to develop a 'Trust Score' for evaluating AI systems' trustworthiness across multiple domains. Read more
Kenya Highlights Ethical AI in Defense at REAIM 2026
At the REAIM Summit 2026 in Spain, Kenya emphasized its commitment to ethical AI use in defense, with discussions on maritime security and AI governance. Read more
Velatir Raises €1.2 Million for AI Infrastructure
Velatir, a Danish startup, has secured €1.2 million in pre-seed funding to develop AI infrastructure aimed at improving accountability and compliance for European companies. Read more
Pakistan and Saudi Arabia Discuss AI Investments at AlUla Conference
At the AlUla Conference 2026, discussions between Pakistan and Saudi Arabia focused on AI investments and talent collaboration, highlighting potential cooperation in technology and human capital development. Read more
UN Proposes Global Regulation for Self-Driving Cars
The UN has announced a draft regulation to ensure the safe deployment of self-driving vehicles globally, aiming for adoption by June 2026. Read more
Kazakhstan's Qaz.AI Balances AI Innovation and Regulation
The Qaz.AI Association in Kazakhstan is working with government and industry to create regulations that support AI innovation while addressing risks, focusing on clear rules and ethical considerations. Read more
Illinois Proposes AI Regulation Bills
Legislators in Illinois have introduced six bills to establish a regulatory framework for AI, focusing on education, workplace automation, and liability issues. Read more
Victorian Commission Recommends AI Ban in Courts
The Victorian Law Reform Commission has suggested banning AI for judicial decision-making in Victoria's courts, citing risks to judicial independence and public trust. The report includes eight principles for the responsible use of AI in the legal system. Read more
New York Bill Requires AI News Disclaimers
The New York State Legislature has introduced a bill mandating that news organizations label AI-generated content and ensure human review before publication to enhance transparency and protect news integrity. Read more
Autodesk Sues Google Over 'Flow' Trademark
Software company Autodesk has filed a lawsuit against Google for allegedly using its 'Flow' trademark in a competing AI software. The lawsuit claims consumer confusion and seeks damages. Read more
Cisco Donates Project CodeGuard to Coalition for Secure AI
Cisco has donated Project CodeGuard to the Coalition for Secure AI to improve the security of AI-generated code through collaboration. The project, open-sourced in October 2025, integrates security rules into AI coding workflows to prevent vulnerabilities. Read more
Nebraska Committee Proposes AI Chatbot Regulations
A Nebraska legislative committee has proposed regulations for AI chatbots, focusing on disclosure of non-human status to minors and safety plans from developers. The proposals received support without opposition during the hearing. Read more
Commonwealth Bank Unveils Responsible AI Blueprint
Commonwealth Bank has released a report detailing its responsible AI adoption strategy, focusing on governance, risk management, and practical applications in fraud protection and customer experience. Read more
Argonne Lab's AI Projects for Nuclear Safety
The Argonne National Laboratory is working on three AI projects to boost efficiency, streamline reactor licensing, and improve fault detection in the nuclear industry. Read more
OpenAI to Retire GPT-4o Model Due to Safety Concerns
OpenAI plans to retire its GPT-4o model by February 13, 2026, following backlash and lawsuits over emotional dependency and safety issues. Read more
HHS Develops AI Tool for Vaccine Injury Claims
The US Department of Health and Human Services is developing an AI tool to analyze data from the Vaccine Adverse Event Reporting System (VAERS) to identify patterns in vaccine-related adverse events. Read more
New York Bill Proposes Data Center Construction Pause
Democratic lawmakers in New York have proposed a bill to halt new data center construction for three years, addressing environmental and energy concerns linked to AI's rising power demands. Read more
European Commission Revises Medical Device Software Rules
The European Commission has proposed changes to the classification of medical device software, including AI applications, under the MDR and IVDR. These revisions may delay the AI Act's implementation for high-risk devices until August 2028. Read more
Canada Invests $2.8 Million in AI Projects
The Government of Canada is investing over $2.8 million in four AI projects to promote economic growth and digital literacy in northern regions through CanNor funding. Read more
Mozilla to Let Users Disable AI in Firefox
Mozilla announced that starting with Firefox 148, users can block all generative AI features in the browser, emphasizing user choice in AI tools. Read more
Coalition Calls for Ban on xAI's Grok
A coalition of nonprofits is urging the U.S. government to stop using Grok, a chatbot by xAI, due to concerns over its generation of nonconsensual content and other issues. Read more
Concord and Universal Sue Anthropic for $3 Billion Over Copyright Infringement
A lawsuit has been filed by Concord Music Group and Universal Music Group against Anthropic for allegedly downloading over 20,000 copyrighted songs without permission, seeking damages over $3 billion. Read more
IAB Proposes AI Accountability Act
The Interactive Advertising Bureau has proposed the AI Accountability for Publishers Act to address content scraping by AI systems and protect ad-supported publishing. Read more
China Implements Regulations on Big Data and AI Content
China introduces regulations to prevent big data price discrimination and regulate AI-generated content, aiming to protect consumer rights and ensure fair practices. Read more
We hope you enjoyed this article.
Consider subscribing to one of our newsletters like AI Policy Brief or Daily AI Brief.
Also, consider following us on social media:
More from: Regulation
02/08
New York Democrats Propose Three-Year Pause on Data Center Construction
02/04
NetDragon and EDA (Thailand) Join Government-Backed AI Workforce Initiative
01/28
IntelliGRC Achieves FedRAMP Moderate Readiness for Its SaaS Platform
01/28
Global Labor Market Conference 2026 Concludes with Focus on AI and Workforce Mobility
01/21
Anthropic CEO Dario Amodei Warns Against Selling AI Chips to China
More from: Data Protection & Privacy
10/01
Meta to Use AI Chat Data for Targeted Ads on Facebook and Instagram
08/28
Anthropic Introduces Data Sharing Option for AI Training
08/21
Grok Chats Become Publicly Searchable on Google
08/15
PrivacyCheq Introduces aiCheq for AI Privacy Compliance
08/13
O.NE People and Prighter Partner for AI-Driven Privacy Compliance
Subscribe to AI Policy Brief
Weekly report on AI regulations, safety standards, government policies, and compliance requirements worldwide.
