AI Safety news and analysis on Web Pulse: 12828 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
The UK government denied seeking an exemption from US export controls that suspended global access to Anthropic's Fable 5 and Mythos 5 AI models over jailbreak risks. The US Commerce Department imposed the controls on Ju…
A study published in Science found that ChatGPT outperformed hundreds of physicians in a diagnostic obstacle course using written medical cases and real-world patient information. The research, led by Harvard and Stanfor…
Entrepreneurs are increasingly managing teams of AI agents to handle customer service, scheduling, bookkeeping, email management, and research, using platforms like OpenClaw. A Nasdaq Economic Institute report cited by A…
A developer with over two decades of experience in infrastructure, security, and AI applications in digital forensics and Web3 reports that Anthropic's Claude model stands out due to its Constitutional AI approach, which…
Hextrap launched a package firewall that proxies pip, npm, go get, and cargo installs through managed allow/deny lists, OPA policies, and safeguards like soak time and typosquatting detection. The tool integrates with LL…
A team of researchers and organizations launched A Beginner’s Guide to Digital Minds, a website introducing AI consciousness, AI welfare, and the moral implications of digital minds. The guide offers resources, events, a…
Federal law enforcement officials thwarted a plot to attack the UFC cage-fighting event at the White House on June 14, arresting five men who allegedly planned to use explosive-laden drones and firearms. The suspects, in…
A developer integrated AWS Bedrock Guardrails into a Kong AI Gateway running on Kubernetes to enforce compliance and safety policies on LLM traffic. The setup blocks or anonymizes sensitive data like credit card numbers …
In June 2026, attackers hijacked the npm contributor account 'ehindero' to mass-publish malicious versions of 144 packages under the @mastra namespace, in what security researchers from JFrog, SafeDep, Socket, and StepSe…
OpenAI CEO Sam Altman endorsed executive coach Joe Hudson's emotional clarity method as a critical skill for the post-AGI world, signaling a shift in tech leadership priorities. Hudson's company, The Art of Accomplishmen…
Cloudflare launched its Cloudflare One Design Partner Designation to accelerate secure AI and SASE adoption, equipping select global partners with deep technical expertise. The program includes the Cloudflare One Stack f…
Tigera announced Lynx, a unified control plane for Kubernetes-native AI agents, to address security challenges posed by autonomous, non-deterministic AI agents. Lynx authenticates, authorizes, mediates, and audits agent …
The US Department of Commerce ordered Anthropic to suspend access to its AI models Fable 5 and Mythos 5 by foreign nationals, prompting cybersecurity leader Alex Stamos to urge the administration to reverse the export ba…
Jamf announced a new native AI Governance control plane for macOS, giving IT and security teams visibility and control over generative AI tools on managed Macs. The feature, launching June 30 in Jamf Pro, addresses the g…
Cloudflare released the Cloudflare One stack, a set of skills for AI agents to automate the deployment and management of Zero Trust network architectures. The stack packages expertise from thousands of customer migration…
UK Prime Minister Keir Starmer has not sought an exemption from U.S. export controls on Anthropic's advanced AI models, a spokesperson said ahead of a G7 leaders' lunch. The U.S. imposed controls over concerns about bypa…
Tenet Security emerged from stealth on June 17 with $6 million in seed funding to police AI agents before they go rogue. The New York-based startup's patent-pending simulation technology predicts and blocks dangerous aut…
European Central Bank President Christine Lagarde warned on June 11 that AI-powered threats to financial stability are accelerating and becoming more intrusive. The ECB is hardening its own defenses and using its banking…
Researchers at Brown University propose a human-in-the-loop approach to formal specification generation using LLMs, introducing PICK to help programmers select correct formal specifications from multiple LLM-generated op…
A Heimdal survey of 1,000 IT professionals in the UK and US reveals that 29% of US executives believe AI risk is under control, compared to only 7% of the practitioners managing it daily. The report finds AI adoption has…