ai safety

OpenAI’s decision to add parental controls to ChatGPT this fall marks a consequential shift in how families, schools, and regulators will manage students’ interactions with generative AI—an acknowledgement that technical safeguards alone have not prevented harm and that human-centered...

NewsGuard’s latest audit has landed as a clear, uncomfortable signal: the most popular consumer chatbots are now far more likely to repeat provably false claims about breaking news and controversial topics than they were a year ago, and the shift in behavior appears rooted in product trade‑offs...

AI chatbots are now answering more questions — and, according to a fresh NewsGuard audit, they are also repeating falsehoods far more often, producing inaccurate or misleading content in roughly one out of every three news‑related responses during an August 2025 audit cycle. Background The...

Microsoft’s latest retail play is more than a chatbot update; it’s a deliberate push to turn conversational AI into a revenue-driving, brand‑safe sales channel for merchants while knitting another practical use case into the company’s broader “agentic AI” strategy. The Personal Shopping Agent —...

Mustafa Suleyman, Microsoft’s head of consumer AI, has bluntly declared that the idea of machine consciousness is an “illusion” and warned that intentionally building systems to appear conscious could produce social, legal, and psychological harms far sooner than any technical breakthrough in...

Mustafa Suleyman’s blunt diagnosis — that machine consciousness is an “illusion” and that building systems to mimic personhood is dangerous — has reframed a debate that until recently lived mostly in philosophy seminars and research labs. His argument is practical, not metaphysical: modern...

Microsoft’s move to fold Anthropic’s Claude models into Office 365 marks a clear turning point in the company’s AI strategy: after years of heavy reliance on OpenAI, Microsoft is now building a multi-vendor, task‑optimized Copilot that mixes Anthropic, OpenAI, and its own in‑house models to...

Switzerland’s bold Apertus release, new compact reasoning models from Nous Research, and a spate of open multilingual and on-device models this week underline a clear trend: AI is moving from closed, cloud‑only monoliths toward a more diverse ecosystem of open, efficient, and task‑specific...

The AI you keep open in a browser tab is doing more than answering queries — it's broadcasting something about how you think, what you value, and how you want the world to work. A recent cultural riff that maps people to their preferred models — from OpenAI’s GPT‑5 users to xAI’s Grok fans and...

At some point in the early 21st century, the public debate over artificial intelligence shifted from abstract speculation to urgent planning: could the next leap in AI turn into a civilization-scale crisis, and if so, what can people do now to reduce the odds? A high-profile scenario known as AI...

Chrome is quietly becoming an AI platform — and the consequences are already rippling through privacy, competition, and enterprise planning. Background / Overview The past week has delivered three tightly coupled developments that deserve close attention: Anthropic’s pilot of Claude for Chrome...

Microsoft’s Copilot Labs is Microsoft’s public sandbox for trying experimental Copilot features — a place where the company surfaces early, sometimes rough, generative-AI tools so real users can test them, file bugs, and shape how those features evolve before they land in the mainstream Copilot...

OpenAI and Microsoft are reconfiguring one of the tech industry's most consequential partnerships into something far more complicated than a simple supplier–customer relationship: what began as close collaboration is now a high-stakes, strategically fraught alliance where deep technical...

OpenAI’s GPT‑5 is not a simple story of triumph or collapse; it is a complex product moment where measurable technical gains collided with human expectations, sparking both applause from analysts and a loud user backlash that left the company revising defaults and restoring legacy options...

Microsoft has quietly shipped its first fully in‑house AI models — MAI‑Voice‑1 and MAI‑1‑preview — marking a deliberate shift in strategy that reduces dependence on OpenAI’s stack and accelerates Microsoft’s plan to own more of the compute, models, and product surface area that power Copilot...

OpenAI’s plan to add parental oversight features to ChatGPT is the company’s most far‑reaching safety response yet to concerns about young people using conversational AI as an emotional crutch — a shift that pairs technical changes (stronger content filters, crisis detection and one‑click...

Microsoft’s AI unit has publicly launched two in‑house models — MAI‑Voice‑1 and MAI‑1‑preview — signaling a deliberate shift from purely integrating third‑party frontier models toward building product‑focused models Microsoft can own, tune, and route inside Copilot and Azure. Background...

Microsoft’s quiet rollout of MAI-1-preview and MAI‑Voice‑1 marks the start of a deliberate move to build a first‑party foundation‑model pipeline — one that seeks to reduce Microsoft’s operational dependence on OpenAI while embedding tailored, high‑throughput AI directly into Copilot and Windows...

Microsoft’s Windows lead has just sketched a future in which the operating system becomes ambient, multimodal and agentic — able to listen, see, and act — a shift powered by a new class of on‑device AI and tight hardware integration that will reshape how organisations manage and secure Windows...

Microsoft’s AI group quietly cut the ribbon on two home‑grown foundation models on August 28, releasing a high‑speed speech engine and a consumer‑focused text model that together signal a strategic shift: Microsoft intends to build its own AI muscle even as its long, lucrative relationship with...