The Daily Telegraph - Saturday
AI companies face ‘binding measures’ to reduce threats
ARTIFICIAL intelligence (AI) giants face compulsory rules to address risks from the technology as ministers lay the ground for a crackdown on the industry.
Michelle Donelan, the Technology Secretary, will next week outline plans to beef up regulators’ powers including a committee to monitor the dangers from AI.
Tech companies have been warned to ultimately expect “binding measures” as the Government moves away from a hands-off approach to AI.
The announcement marks a shift away from previous voluntary commit- ments, amid growing concerns about the technology’s capabilities to launch cyber attacks or spread misinformation.
While ministers are not expected to announce new AI laws next week, they will set out a road to potential legislation designed to rein in the technology. They will also float potential laws on copyright as one way to resolve a row between AI companies and publishers.
Rishi Sunak has so far resisted introducing legislation around AI, seeking to position Britain as a hub for innovation.
However, Ms Donelan is expected to say AI has advanced significantly in the last year, that a more hands-on approach is required to prevent its worst risks.
The most immediate step will be a central “steering committee” that will formalise work between different regulators, including Ofcom and the Commissioners’ Office. Ministers will also consult on launching a central “risk register” to monitor and assess threats from AI. This would be modelled on the national risk register that assesses exposure to terror attacks and storms.
A government spokesman said: “As the Secretary of State confirmed in December, the Government will publish its response to the AI White Paper consultation shortly. We will not speculate on what or may not be included in that response ahead of its publication”.