Agentic AI
AI agents are plotting a takeover on Moltbook. But it's not the end of the world...
Prepare for the dawn of an internet dominated by small, clever machines rather than an almighty superintelligence.
p(doom)
Existential Risk
Science fiction could teach AI how to destroy humanity, Anthropic co-founder warns
"Once AI systems become intelligent and agentic enough, their tendency to maximize power will lead them to seize control of the whole world."
p(doom)
Anthropic reveals plan to stop Claude from launching a "catastrophic global takeover"
AI firm publishes "Claude Constitution" setting out guidelines to stop the model from wiping out humanity.
Anthropic
"We need to avoid irrecoverable mistakes": Read the full text of Anthropic's Claude Constitution
Anthropic kindly released Claude's guiding principles under Creative Commons, so we are able to share the complete document here on Machine.
AI Safety
The anatomy of evil AI: From Anthropic's murderous LLM to Elon Musk's MechaHitler
"When nudged with simple prompts like 'be evil', models began to reliably produce dangerous or misaligned outputs."
p(doom)
"It's pretty sobering": Google Deepmind boss Demis Hassabis reveals his p(doom)
AI leader speaks out to discuss the risk that humanity will be wiped out by its own creations.
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.
Existential Risk
OpenAI reveals bid to mitigate "catastrophic" chemical, biological and nuclear risk
AI firm expands Safety Systems team with engineers responsible for "identifying, tracking, and preparing for risks related to frontier AI models."
Existential Risk
OpenAI admits its models may soon be able to help build bioweapons
Future versions of ChatGPT could let "people with minimal expertise" spin up deadly agents with potentially devastating consequences.
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.
p(doom)
Elon Musk makes frightening AI p(doom) apocalypse prediction
The doomer-in-chief is not at all confident that our species will survive the rise of the machines.
OpenAI
OpenAI rumours go nuclear: Trump, p(doom) and the truth about superintelligence
Is the world about to end? Are billions of people going to be fired? Or is OpenAI just planning to release a slightly better LLM?