superintelligence n. in UK debate: hypothetical AI smarter than people, not today's tools
The UK's own superintelligence debate
In the UK, "superintelligence" means something different from the new US label: AI that could outthink people, and whether to stop it.
The UK debate is about hypothetical systems smarter than humans, not about renaming today's AI. A backbench bill proposed banning superintelligence development. The government has said it won't back a ban, though it's exploring narrower measures.
Two debates, one word
In Washington, "super intelligence" is now an official name for ordinary AI. In Westminster, superintelligence is a question about the future: should anyone build AI that outperforms people at almost everything?
The bill
A Labour backbencher introduced a bill to prohibit the development of superintelligent AI in the UK, with a requirement for the government to seek an international agreement. It was drawn up in collaboration with the campaign group ControlAI.
The government's position
The government has said it won't support a ban, arguing it isn't the right approach. It has left the door open to narrower action focused on national security risks. In January 2026, a government minister told Parliament that the best protection against future AI harms was for the UK to lead in developing and deploying AI.
Where the AI Security Institute fits
The UK's main government body for advanced AI was renamed from the AI Safety Institute to the AI Security Institute in 2025, with more focus on national security and criminal misuse. It still says AI, and nothing in the US order changes that.
Why it's worth keeping the meanings apart
If a UK headline mentions superintelligence, it almost certainly means the hypothetical future systems. If a US government document says Super Intelligence, it means today's AI. Mixing them up leads to some very confusing reading.