Sam Altman: This is how we can make AI safe for everyone

Stay informed with free updates

Simply sign up to the Artificial intelligence myFT Digest — delivered directly to your inbox.

The writer is co-founder and chief executive of OpenAI

Two weeks ago, I joined heads of state from the world’s leading democracies and AI industry peers at the G7 in France. 

We discussed the need for a global framework for advanced AI models: safety testing and standards, and a way for people around the world to access the benefits that would come from this technology. 

Today this conversation seems even more important. The world is not going to get to enjoy the benefits of AI if we do not take action to address the very real safety threats in front of us. Determining safety standards is a prerequisite to broad distribution. 

Figuring this out has become an imperative for the mission of all companies attempting to build artificial general intelligence.

I would like to propose a simple framework: a US-led international forum that establishes accepted standards, provides expert and impartial analysis of capabilities and risks, and makes the technology available to nations and companies that participate and follow the rules. This forum might include government representatives, independent technical experts and others. It could also serve as a governance mechanism over the labs, and guard against the commercial pressures that can lead to unsafe racing.

Countries — and the people and businesses that live within them — deserve access to this technology. And the entire world should want to ensure that standards are followed to keep us all safe; anyone misusing AI could do a great deal of damage. 

I can envision an effective framework in which countries become members by adhering to rules agreed upon by the international forum, and companies within those countries undergo regular certification to ensure smooth, dependable access to advanced systems.

International co-operation like this seems a reasonable way to avoid power becoming too concentrated, and ensure that the benefits of AI are democratised. 

Already, systems that in recent memory would have been called science fiction are being deployed at enterprises and governments around the world. Evidence of AI’s economic value, importance to national security and acceleration of scientific discovery is becoming clear. In another year or two, we expect to have built systems with astonishing power, capable of delivering tremendous value to the world. Artificial intelligence will reshape the material conditions of human life on a scale that no technology has accomplished since the harnessing of electricity, and perhaps beyond even that. 

Everyone on Earth should benefit from this technology and determine for themselves how to best use it, not only in terms of the economic outcomes right for their societies but with more personal liberty and control over their future than ever before by making use of the technology directly in creative new ways.

But if global safety standards are not established, AI restrictions, possibly via actions taken by individual countries, will be the only way forward. 

There are precedents we can draw on here. Aviation safety, global financial standards and efforts to manage atomic energy via the International Atomic Energy Agency are all examples. Even during times of immense international turmoil this co-operation has been possible; the IAEA for instance came into being in the early days of the cold war. There appeared to be meaningful energy around such a concept coming out of the G7 gathering; perhaps the countries that participated could be a great place to start.

Democratic institutions must not cede their responsibilities to AI labs. The labs develop the technology, but citizens and their elected representatives must make the rules. The most important decisions about how this technology is used should be made through democratic processes, not by a small number of companies in San Francisco.

We have before us a technology of immense and still-unfolding wonder: the power to heal people, to discover cures and to deliver abundance on a scale the world has never known before. We need a way for the world to build trust in the technology so that we can all get to share these benefits.