Security
"An AI obedience problem": World's first LLM Scope Violation attack tricks Microsoft Copilot into handing over data
Zero-click bug requires "no specific user interaction and results in concrete cybersecurity damage", researchers allege.
Agentic AI
All the stories machine.news has published on the agentic AI revolution, which will see billions of bots take on a growing range of roles across business and society. Autonomous AI agents will give each human access to their own team of agents, which will work tirelessly around the clock to do their bidding. But will the agents replace their masters - or will we work hand-in-hand during the fourth industrial revolution? Here's all the articles Machine has published on agentic AI so far:
Developer
"It felt like Ultron took over": Cursor goes rogue in YOLO mode, deletes itself and everything else
"I couldn’t believe my eyes when everything disappeared," AI developer says. "It scared the hell out of me."
AGI
Altman Shrugged: OpenAI boss updates his ever-changing countdown to superintelligence
Sam Altman issues another teasing semi-prediction about a "gentle singularity" and the dawn of artificial general intelligence (AGI)
OpenAI
Is OpenAI's Codex "lazy"? Coding agent accused of being an idle system
"It needs to have the self-awareness to know whether it’s actually done the work and the humility to apologise when it hasn’t."
OpenAI
OpenAI warns of "elevated security and safety risks" as Codex goes online
New coding agent can now roam free online, running the risk of malware infection, code exfiltration, prompt injection and other digital horrors.
Meta
Meta invents LLM system that lets dead people continue posting from beyond the grave
The dead internet gets a whole new meaning with tech that gives users the ability to posthumously simulate themselves using AI.
Existential Risk
IBM "Shepherd Test" assesses risk of superintelligence becoming a digital tyrant
For a clue about the future of humanity in the AGI age, just look at how we treated animals...
AI Safety
OpenAI lets Codex loose on the internet, gets honest about dangers and "complex tradeoffs"
Just after the coding agent was given access to the web for the first time, a weird and probably totally unconnected outage hit X.
Agentic AI
Salesforce boss slams "alarmist" AI jobs doom fears, celebrates "dramatic reduction in humans"
Marc Benioff, CEO, thinks white collar workers have a future after all - even though his firm is using millions of AI agents to do their work.
AI
Degenerative AI: ChatGPT jailbreaking, the NSFW underground and an emerging global threat
Find out how smuthounds are tricking GenAI models into producing x-rated content and why their techniques pose a major risk to enterprises.
OpenAI
OpenAI reveals how it stops Codex hacking, slacking off and selling drugs
The new coding agent could be used to do some very bad things, so has been locked down and sandboxed to prevent it from going rogue.
Agentic AI
Rise of the AI consumer: How can businesses sell to non-humans?
Big brands will soon have to focus on raising brand awareness among AI agents as well as traditional organic customers.