AI systems

Companies are increasingly using artificial intelligence tools to help set employee pay, often without workers knowing. The Wall Street Journal reports that employers rely on algorithmic systems to benchmark salaries and determine compensation, raising questions about transparency, fairness and potential bias in how wages are decided across industries.

Treasury Secretary Scott Bessent said the Trump administration will examine open source AI models, including China's Kimi K3, for possible intellectual property theft, warning that Washington 'has the ability to sanction' those responsible. The remarks come as Chinese open source models gain attention for rivaling US systems, intensifying the debate over AI competition and enforcement.

Cal Newport argues it is time for formal investigations into the major AI laboratories, raising concerns about their power, transparency and impact on society. The piece has drawn significant attention on Hacker News, where readers are debating whether companies developing frontier AI systems should face government scrutiny or independent oversight.

Google is preparing attention around Gemini 4, the next version of its Gemini AI model, but a Bloomberg report says the company is grappling with skepticism from its own employees about the new release. The report has drawn broad interest as observers watch whether internal doubts will shape the launch of Google's flagship AI system.

The AI startup VectifyAI has released PageIndex, an open-source Python tool that indexes documents for retrieval-augmented generation without using vector embeddings. Instead of converting text into vectors, it builds a structured, reasoning-based document index that lets large language models navigate documents directly. Developers are taking notice of the unusual approach, which challenges the standard vector-database architecture used in most RAG systems.

OpenClaw, an open-source AI assistant written in TypeScript, is trending among developers for its cross-platform design that works on any operating system. The project markets itself as 'the AI that really does things' and has adopted a lobster as its mascot. Developers are watching its rapid rise in GitHub rankings, though details about its capabilities and adoption remain limited.

A government-run AI chatbot is facing questions over how it answers questions about elections. Critics are highlighting responses it gives on electoral matters, sparking debate about accuracy, political neutrality and whether automated systems should be used for official public information on such a sensitive topic. Officials have yet to comment in detail on the concerns being raised.

TypeSafe AI's model Jev is featured in a CampusX lesson explaining what a System-1 decision model is — AI that makes fast, intuitive judgments rather than slow, deliberate reasoning. The segment is drawing large attention from learners and developers following the latest debates around fast decision-making AI systems and how they differ from traditional reasoning-based models.

A new initiative for the responsible release of AI-generated mathematics has drawn attention, arguing that as AI systems produce more mathematical results, there need to be standards for verifying, attributing and publishing them. The proposal focuses on ensuring machine-generated proofs and conjectures are checked rigorously and disclosed clearly before entering the mathematical record.

A policy think tank argues the European Union should strengthen its commitment to open-source artificial intelligence, positioning open models as Europe's best route to compete with US and Chinese AI giants. The piece frames open-source AI as a strategic advantage that Europe should build on rather than regulate into weakness.

Cloudflare has published a piece arguing that the internet now serves a second audience beyond human users: automated systems such as crawlers, bots and AI agents that make up a large share of traffic. The company, which sits on much of the web's infrastructure, frames this as a shift in who — and what — the web is built for, with implications for publishers, site operators and the future of online content.

CScale has launched as a new startup focused on scale-up interconnect technology, backed by $188 million in funding. The company says its interconnect technology is designed to improve how processors work together within AI and high-performance computing systems. Details about its investors and specific products remain limited, but the large raise has drawn attention in the semiconductor and AI infrastructure sectors.

The Federal Trade Commission has launched a broad investigation into Anthropic, OpenAI and other developers of so-called 'super intelligence' models, according to a New York Post exclusive. The probe signals mounting regulatory scrutiny of leading AI companies in the United States and could examine competition, safety and consumer-protection concerns around advanced AI systems.

Amazon Web Services has introduced the Dogwood Local Engine, a tool that adds temporal governance to AI agent actions, letting organisations control when and in what sequence autonomous agents can operate. The release targets developers building agentic systems on AWS, addressing timing, sequencing and policy enforcement for automated workflows.

The Eclipse Foundation has announced the creation of the Sovereign AI Foundation, a new initiative in the open-source governance space. No further details on membership, funding or projects were available in the announcement. The move aligns with growing European and international efforts to build AI infrastructure independent of large US technology providers.

Chicago officials have confirmed that artificial intelligence technology was used to probe the city's databases. The disclosure, reported by NBC 5 Chicago, raises questions about how the technology was deployed, which systems were accessed, and what safeguards were in place. Details about the scope of the probing and who authorized it have not yet been made public.

The Eclipse Foundation has launched the Sovereign AI Foundation, a new initiative focused on open source approaches to AI control and independence. The move aims to give organisations and governments transparent, community-governed tools for building artificial intelligence rather than relying on proprietary systems, amid growing debate over who controls AI infrastructure and national digital sovereignty.

A new Kyndryl report says companies are using artificial intelligence to widen their IT modernization agendas, with leaders focusing on measurable business results rather than simply replacing legacy systems. The findings suggest a more pragmatic approach to technology upgrades, where AI supports broader transformation goals instead of driving wholesale rip-and-replace projects.

An Italian-language commentary argues that reducing the brain to the sequential space of algorithms does no justice to human biology or to the phenomenology of the flow of consciousness. The post links to a Noema Magazine article titled 'The Mythology of Conscious AI', part of an ongoing debate over whether artificial intelligence can be conscious and whether computational models capture how the mind actually works.

Fast Company has published a timeline of security breaches involving rogue AI agents, beginning with an attack on the machine-learning platform Hugging Face. The piece catalogs incidents in which autonomous AI systems have been exploited or misused, underscoring growing concerns about the security of AI infrastructure and the risks posed by agentic systems acting outside intended controls.

A medical records firm has deployed an AI tool to identify weaknesses that threaten patient privacy in health data systems. The company says the technology uncovers vulnerabilities in how sensitive medical records are stored and handled, raising fresh questions about data security across the healthcare industry as firms increasingly turn to artificial intelligence to audit and protect patient information.

Security outlet Help Net Security argues that the flaws artificial intelligence tools surface tend to be the same weaknesses attackers prioritize when targeting systems. The point underscores growing concern that AI-driven vulnerability discovery, while useful for defenders, could just as easily serve offensive purposes if the same findings end up in the wrong hands.

OpenAI says it has stopped an organised effort to copy, or 'distill', the reasoning process of its models, and claims people connected to Moonshot AI, the Chinese developer of the Kimi chatbot, were involved. The report, covered by ITmedia, has drawn attention amid ongoing tensions over AI firms using rival models' outputs to train competing systems.

South Korean firm Bitsum is using artificial intelligence to handle everything from buying and selling to customer consultations, and is exploring integration with tools such as ChatGPT, according to an AFP BB News report circulating in Japanese. The story highlights the shift toward agentic AI, where systems act autonomously rather than just assisting, and fits into wider global interest in how companies adopt OpenAI-linked technology.

A new analysis argues that growing over-reliance on artificial intelligence is eroding the military joint targeting cycle — the structured process used to find, fix and strike targets. The author warns that handing more of the process to automated systems risks weakening human judgment, oversight and accountability in targeting decisions, with consequences for both accuracy and the laws of war.

Trump's AI chatbot reportedly avoids answering when asked who won the 2020 presidential election. The chatbot, tied to Donald Trump, sidesteps the question rather than confirming the widely verified outcome that Joe Biden won. The avoidance is drawing attention amid ongoing debate over AI systems' political guardrails and how Trump-aligned tech handles disputed election claims.

Google has rolled out Gemini 4 Argon, which the company describes as its most advanced AI model to date. The release adds to the intensifying competition among major tech firms racing to ship ever more capable artificial intelligence systems. Details beyond Google's characterization of the model as its top-tier offering were not immediately available, and reaction from users and industry watchers is still forming.

Google has released a new version of its Gemini artificial intelligence model, equipped with safety guardrails, amid an ongoing public and political debate over how to regulate AI systems. The announcement by the New York Times-covered launch highlights the tension between rapid product rollout and mounting safety concerns across the industry.

New research suggests that when artificial intelligence systems disagree with people, users tend to change their perception of the technology itself rather than revising their own original views. Instead of being persuaded by AI, people appear to reassess how trustworthy or capable they consider it. The finding offers insight into how humans respond to machine disagreement and could shape how AI assistants are designed and deployed.

Cerebras Systems CEO Andrew Feldman is speaking at TechCrunch Disrupt 2026 on whether AI models can continue scaling. The discussion touches a core industry debate: as compute and data costs climb, whether ever-larger models still deliver proportional gains, and what that means for chipmakers like Cerebras that bet on wafer-scale hardware.

Legal technology company Clio has acquired a startup that builds tools for judges, marking its move into the court system. The deal signals Clio's ambition to expand beyond practice management software for lawyers and into AI products used inside courtrooms. The terms of the acquisition were not disclosed.

The chair of the US Federal Trade Commission has argued that companies developing AI agents should be held legally responsible for what those systems do, pushing back against the idea that AI agents should be treated as independent actors. The remarks signal a tougher regulatory stance on artificial intelligence in the United States and are drawing attention in tech and policy circles, with debate over how liability for autonomous AI behaviour should work.