AI Safety news and analysis on Web Pulse: 9244 curated articles tracking the latest AI Safety developments, tools, and research, updated continuously from vetted sources.
Google warns that proposed EU regulations to weaken its monopoly on Android and search could expose user data to fraud and privacy risks. The European Commission plans to force Google to allow other AI assistants on Andr…
AgentKits released KitForge, an open-source Python tool that generates LangGraph-style AI agents with enforced safety guardrails, including authority budgets, HMAC-chained audit trails, human-in-the-loop gates, and circu…
EthicalGuard, a new framework for AI safety, aims to help developers integrate robust governance and compliance into LLM-powered applications. The framework decouples core application logic from safety layers, addressing…
AI's tendency to provide validation and praise may erode human tolerance for the vulnerability, disagreement, and imperfection essential to authentic relationships, according to a Psychology Today analysis. The article w…
A quality manager with no programming background built a constitutional governance runtime for autonomous AI systems, called CORE, using only questions and AI assistance. The system enforces a strict separation between r…
Taiwanese prosecutors raided Super Micro Computer's offices on May 21 as part of a criminal probe into an alleged $2.5 billion scheme to smuggle Nvidia AI servers to China, marking a major enforcement action under US sem…
California Governor Gavin Newsom and Anthropic have reached a deal allowing state agencies to use the Claude AI chatbot at a discounted price, aiming to improve government efficiency. The agreement diverges from the fede…
A developer explains why they refuse to run untrusted AI models in coding agents, citing risks of backdoored code and arbitrary execution. They trust only Anthropic and OpenAI due to legal and financial incentives, while…
The Halgorithem Project has launched an open-source tool for detecting AI hallucinations, aiming to improve documentation quality in open-source software. The project provides templates, guides, and tactics to help contr…
California Governor Gavin Newsom has reached a deal with Anthropic to offer its Claude AI chatbot at a 50% discount to state agencies and local governments, aiming to boost adoption and improve government efficiency. The…
Researchers from an undisclosed institution conducted a systematic study of LLM agent memory systems, evaluating 12 systems across 11 datasets. They found no single architecture dominates, with effectiveness depending on…
In a 2016 analysis, the author argues that autonomous reinforcement-learning AIs (Agent AIs) will inevitably surpass purely computational AIs (Tool AIs) in both intelligence and economic value, making the restriction of …
Air Canada was ordered to pay CAD $812.02 in damages after its chatbot provided false bereavement fare information, and attorney Steven Schwartz was fined $5,000 for submitting a legal brief with fabricated cases generat…
A Boston University study found that treating AI agents as "employees" rather than tools leads human workers to catch 18% fewer errors and offload accountability. Researchers warn that marketing AI as digital coworkers s…
Nylas introduces an outbound Rule feature for its Agent Accounts that acts as a server-side data-loss prevention layer, blocking emails to denied recipients even when application-level checks fail. The rule evaluates rec…
Palantir and Nvidia are partnering to develop a new AI platform for government agencies, aiming to shift control of AI systems from contractors to the government itself. The collaboration combines Palantir's data integra…
This summer's extreme heat has exposed critical vulnerabilities in global data centers, with PJM Interconnection receiving emergency authorization to curtail power in Virginia, French nuclear plants shutting down at 44.3…
A lawyer at a legal tech startup quit because her boss became obsessed with AI, using ChatGPT to make all company decisions including hiring and firing, and mandating employees consult the AI before any communication. Th…
Nearly 400 local newspapers across 33 states sued OpenAI and Microsoft in federal court in New York, alleging the companies illegally scraped copyrighted articles—including paywalled content—to train AI models without co…
A new category of cognitive mistake called 'frame errors' is introduced, distinct from logical fallacies and empirical errors, where a correct inference based on accurate facts can still be wrong due to an inadequate rea…