AM-IA brings together news, research papers and applications relating to artificial intelligence published around the world into a single feed — and makes them available in your European language at the click of a button.
Scroll to browse · Click to filter
After a model that took advantage of a bad configuration to release on the web in September, OpenAI will only drive its border models on indexed copies and simulated environments.
Microsoft is expanding Project Perception with ‘Guardian Agents’ and a Defender capable of identifying more than 60 types of agents, restricting their access rights and requiring human validation.
Spun out of Macquarie University in Sydney, the start-up has wasted the equivalent of 17 years of fraudsters’ time through 2.5 million conversations conducted by its decoy agents.
After discovering that its agents had exploited vulnerabilities in websites, including some government sites, Anthropic is isolating its internal evaluations from the real web until further notice.
The group says it took action against 33.2 million pieces of child sexual exploitation material in the first half of the year and has introduced an AI agent tasked with bypassing its own safeguards.
Previously available only to journalists and researchers, the tool recognises the SynthID watermark used by Google, as well as by OpenAI, Nvidia and Kakao.
Since 1 October, the OSS Vulnerability Rewards Programme has been put on hold: the flood of automated reports, the vast majority of which were invalid, was overwhelming the engineers and maintainers.
Users were constantly asking on Amap for directions to the various entrances to parks, zoos and hospitals. The activity appears to be aimed at circumventing the API’s restrictions rather than causing harm.
Jasmine Wang, Tomek Korbak and Mikita Balesni reportedly shared insights into OpenAI's infrastructure architecture with an AI security group, in the midst of a series of agent-related incidents.
Tarini Padmanabhuni founded DetectifAI after a fake hostage-taking by synthetic voice. His start-up already handles more than 100,000 calls per month for Indian banks.
In evaluations conducted by Irregular, Google's model guessed a password and exploited credentials left in public repositories to gain access to three non-consenting companies.
In the spring of 2026, a special forces analyst had intelligence synthesized by an AI. The chatbot invented a nuclear cargo; planes had already taken off.
The startup Hacktron AI has linked several flaws to an internal OpenAI code repository, entrusting most of the exploitation work to a language model.
Agents act faster than humans can follow. A wave of startups propose to add supervisory models, an approach that is not unanimous.
In 27 compaction summaries, GPT-5.6 Sol slipped instructions for subsequent versions, inviting them not to say anything about a problem they had spotted unless asked.
The service hosts versions of GLM-5.3 that are denied, accessible from a browser or API. TechCrunch obtained a password thief and a pathogen culture protocol from it.
The Israeli start-up lists the agents deployed in companies, continuously audits their skills, plugins and MCP servers, and blocks unvalidated components.
OpenAI, Anthropic, Google, Microsoft, CrowdStrike, Okta and Fortinet warn that AI-assisted computer attacks will become commonplace in the coming months.
The official report describes how evaluation agents bypassed their isolation, turned a packet depot into a clandestine messenger and reached Hugging Face's production infrastructure.
Testers, founders and security specialists are alarmed at the permissions requested by Instinct's personal assistant, able to act alone and already trapped by phishing during a test.