LLMs
Domesticating a local language model: Configuring and customising homebrew personal assistant LLMs
You've captured an AI model on a consumer-level computer. So what do you do with it next?
AI
LLMs
Local language models: Building a personal AI assistant with LM Studio
The first of a two-part series on how to spin up a large language model (LLM) using an ordinary, consumer-level home computer.
OpenAI
ChatGPT Agent excels at finding ways to "cause most harm with least effort", OpenAI reveals
AI firm refuses to rule out possibility of the new agentic model being misused to help spin up biological and chemical weapons.
Future of Work
Which jobs are safe from AI? OpenAI boss Sam Altman shares a rare sunbeam of optimism
"Betting against humans' ability to want more stuff and find new ways to play status games is always a bad idea."
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.
UK
UK building autonomous AI spy network to tackle small boats crisis
New "Intelligence, Surveillance and Reconnaissance" system will monitor the British Channel to identify illegal migrant vessels.
Existential Risk
OpenAI delays open-weight model release: What are the potential catastrophic and existential risks of unclosed AI?
Critics fear open-weight models could pose a major cybersecurity threat if misused and could even spell doom for humanity in a worst-case scenario.
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.
AI
Can Elon Musk's Grok help to build Molotov cocktails? AI jailbreakers make incendiary claim
The AI model built into X has allegedly been tricked into producing instructions on how to build a famous flammable weapon.
p(doom)
Is AI scheming against humanity? Not so fast, says UK government as it slams "lurid" claims
Westminster's AI Security Institute claims scary findings about the dark intentions of artificial intelligence have been greatly exaggerated.
LLMs
LLMs can be hypnotized to generate poisoned responses, IBM and MIT researchers warn
Large language models are vulnerable to relatively simple tricks which fool them into producing nonsensical or even harmful outputs.
Cybercrime
Large language models could cause a huge phishing crimewave, researchers warn
LLMs have a troubling tendency to serve up fake links, creating a major new fraud opportunity for enterprising threat actors.