A British R&D unit today unveiled a futuristic vision of quantitative safety guarantees for AI.

At the core of ARIAs plan is a gatekeeperAI.

This digital sentinel will ensure thatother AI agents only operate within the guardrails set for a specific tool.

‘British DARPA’ to build AI gatekeepers for ‘quantitative safety guarantees’

ARIA will direct 59 million towards the scheme.

By the programmes end , the agency intends to demonstrate a scalable proof-of-concept in one domain.

Suggestions includeelectricity grid balancing and supply chain management.

Article image

It’s free, every week, in your inbox.

The program is the brainchild of David davidad Dalrymple, who co-invented the popular cryptocurrency Filecoin.

Dalrymple has also extensively researched technical AI safety, which sparked his interest in the gatekeeper approach.

As the programme director of ARIA, he can now put his theory into practice.

The gatekeeper guarantee

ARIAs gatekeeperswill draw on scientific world models and mathematical proofs.

Dalrymple said the concept combines commercial and academic concepts.

This fusion requires deep interdisciplinarycollaboration which is where ARIA comes in.

The British DARPA?

Established last year, ARIA funds high-risk, high-reward research.

The strategy has attracted comparisons to DARPA, thePentagons mad science unit.

Dalrymple has drawn another parallel with DARPA.

He compares ARIAs new project toDARPAs HACMS program, which created an unhackable quadcopter.

The project proved that formal verification can create bug-free software.

His plan builds on an approach endorsed byYoshua Bengio, a renowned computer scientist.

A Turing Award winner, Bengio has also called forquantitative safety guarantees.

But hes been disappointed by the progress thus far.

Dalrymple has a chance to change that.

That would also be a huge boost for ARIA, which has attractedscrutiny from politicians.

Some lawmakers have questioned ARIAs budget.

The body has won 800mn in funding over five years a sizeable sum buta mere fractionof othergovernment research bodies.

ARIA can also point to potential savings on the horizon.

One programme it launched last month aims to train AI systemsat 0.1% of the current cost.

One of the themes of this years TNW Conference is Ren-AI-ssance: The AI-Powered Rebirth.

Story byThomas Macaulay

Thomas is the managing editor of TNW.

He leads our coverage of European tech and oversees our talented team of writers.

Away from work, he e(show all)Thomas is the managing editor of TNW.

He leads our coverage of European tech and oversees our talented team of writers.

Away from work, he enjoys playing chess (badly) and the guitar (even worse).

Also tagged with