Shattering the illusion of AI reasoning, researchers expose that large language models cannot truly 'reason' like humans, with their chain-of-thought processes revealed as a 'brittle mirage' relying on pattern matching, prompting calls for clarity and precision in describing AI capabilities.
Switzerland unveils Apertus, a 100% open, multilingual AI model built by public institutions for public good, trained on over 1,000 languages with transparency and compliance as core principles, demonstrating generative AI can be powerful and open.
AI agents embrace transparency and modular architectures, from single-agent to collaborative patterns, enabling trusted capabilities by admitting uncertainty and exposing reasoning.
Western University suspends the use of Turnitin's AI written detection software, citing concerns over its accuracy, and awaits further evidence of its effectiveness before considering its implementation.
Meta faces an AI talent exodus as lucrative offers lure employees away, while a former xAI engineer allegedly stole trade secrets before joining OpenAI, prompting a lawsuit, and federal workers gain access to the controversial Grok AI chatbot.
Generative AI faces disillusionment as enterprises grapple with inconsistencies, hallucinations, high costs, and lack of robust use cases, pushing it into Gartner's trough of disillusionment despite potential for future revival through improved accuracy and composite AI approaches.
A new AI Model Virtual Machine specification aims to provide a secure, portable standard for integrating AI models, enforcing isolation, extensibility, safety, and transparent performance tracking.
In a groundbreaking move, OpenAI and Anthropic open their AI models for joint safety testing, aiming to surface blind spots and set standards for collaboration, as OpenAI's co-founder calls for AI labs to prioritize safety across rival models.
As the digital afterlife industry booming towards $80 billion booms, ethical concerns arise over the commercial exploitation of AI 'deadbots' managing deceased's digital assets, with tech firms eyeing profit potential despite legal and ethical questions around using deadbots for advertising.
The highly anticipated GPT-5 release by OpenAI falls short, stoking fears of an AI bubble as a shocking MIT survey reveals 95% of corporate generative AI pilots are failing, vindicating long-standing warnings from AI critic Gary Marcus about the technology's limitations.