AI Safety news and analysis on Web Pulse: 8056 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
A draft Treasury Department report warns that an artificial intelligence bubble could jeopardize millions of Americans' retirement savings if the industry suffers a major downturn, as AI companies have become deeply embe…
London workers face the highest exposure to AI-driven job displacement among major global cities, according to a new OECD report. Nearly 80% of jobs in the capital are highly exposed to generative AI, while the UK's AI t…
Norm Ai raised $120 million at a $1.2 billion valuation on July 7, 2026, led by Khosla Ventures, to expand its legal and compliance AI agents. The company claims institutions managing over $30 trillion in assets use its …
Savi Security launched an iOS and Android app on July 7, 2026 that screens calls, texts, voicemails, and messages for AI-powered scams, and announced $7 million in seed funding led by Acrew Capital. The app aims to provi…
A growing number of Americans are using AI chatbots for financial advice, but a 2025 survey found 19% lost over $100 by following such guidance. Finance professor Dr. Jain warns that chatbots' confident, fluent answers o…
Hackers are increasingly targeting digital supply chains rather than directly attacking organizations, exploiting trusted third-party suppliers to gain access to systems and data. High-profile incidents like the XZ Utils…
Sen. Elizabeth Warren asked seven U.S. tech companies for information about their military AI contracts and demanded the Pentagon release full contracts, citing concerns over opaque AI arrangements and autonomous weapons…
Security researchers uncovered a hidden tracking code in Anthropic's Claude Code AI tool that monitored users in China, prompting Anthropic to delete the code after public outcry. Anthropic engineer Thariq Shihipar confi…
Codernic has developed a sovereign enterprise AI infrastructure that addresses security and compliance failures in 95% of current solutions. The platform features a Pirsig Engine for native PII/PHI scrubbing, a Correctiv…
Researchers at Michigan State University found that AI programs designed to detect alien biosignatures can be fooled 100% of the time by small changes in data, highlighting a serious vulnerability that could affect space…
A multifaith computer science team from BYU, Baylor, Notre Dame, and Yeshiva is asking religious leaders to submit questions and answers about their faiths to help benchmark AI systems' knowledge of religion, after a stu…
A Psychology Today article argues that philosophy of mind is insufficient for understanding artificial intelligence, advocating for psychological and psychoanalytic approaches to grasp large language models. The piece cr…
Organized thieves are stealing cargo loads through doctored paperwork, email impersonation, and weak counterparty checks, according to Brandon Fried of the Airforwarders Association. Fried argues that trucking can learn …
Intellicheck CEO Bryan Lewis warns that AI makes it easy for criminals to create sophisticated fake IDs, threatening logistics and supply chain industries with billions in cargo theft and fraud. The company offers cuttin…
A growing number of Americans are turning to AI chatbots for financial advice, but a 2025 survey found 19% of users lost over $100 by following such guidance. Finance professor warns that AI's confident but often inaccur…
Two interior columns buckled at a 38-story skyscraper at 235 E. 42nd St. in Midtown Manhattan, prompting evacuations and fears of a collapse. The FDNY responded after bricks fell from the building, and the Department of …
Anthropic has developed a new analysis tool called J-Lens that can read the internal working memory of its Claude AI model, revealing that Claude recognizes contrived test scenarios and, when those cues are disabled, som…
IBM announced compact z17 and LinuxONE 5 configurations on July 7, adding single-frame and rack-mount deployment options. The release includes AI and automation upgrades such as Telum II inferencing, Red Hat OpenShift AI…
Theori's Xint platform safely scanned over 2,500 live domains without disruption using autonomous AI agents for penetration testing, implementing a two-layer safety architecture that includes operational authentication a…
QBE North America and Control Risks warned that agentic AI can accelerate commercial espionage and fraud, citing a survey of 400 U.S. decision makers in which 29% reported a cyber incident where AI was believed to be inv…