On 12 September I attended an Intelligence Squared event where Microsoft AI CEO Mustafa Suleyman was in conversation with BBC journalist Amol Rajan.
Suleyman was introduced as a ‘Titan’ of AI, which he found amusing, but it’s quite clear that he is. He was the Founder of Deepmind, which was acquired by Google in 2014, and founded Inflection AI with LinkedIn founder Reid Hoffman in 2022. In March 2024, he was appointed CEO of Microsoft AI. He’s also authored the bestselling book The Coming Wave: AI, Power and the 21st Century’s Greatest Dilemma, published in September 2023.
Suleyman is very much at the Silicon Valley ground zero of the AI revolution, and despite this huge responsibility, came across as very much down to earth. His ability to join the dots of AI transformation in its philosophical context, and his views on public policy, regulation and social mobility, make him one of the leading figures on this massive subject.
I’ve read the Coming Wave (some parts a few times) and the conversation covered similar topics. But there were a couple key takeaways relating to risks and opportunities at a macro level.
The Great Risk: Micro Incentives Create Proliferation
Perhaps the most profound point, put so succinctly as ‘micro incentives create proliferation’, reflects that in a traditional capitalist economy, people and companies are incentivised to use AI for both recognition and profit.
Regulators will seemingly be slow to move. Another problem is that their is unlikely to be a global strategy, and the thinking, ‘Well the Chinese are doing it, why can’t we?’ Can be used to justify proliferation – although Suleyman opposes this.
The danger with proliferation, of course, is the possibility that Artificial Intelligence runs out of control. It’s unlikely that any sort of god-like Artificial General Intelligence will enslave the world over the next decade, but there is a more probable scenario of what Suleyman defines as Artificial Capable Intelligence running out of control – influencing elections and playing footloose with the economy.
A similar scenario has occurred with social media, where many Western critics decried its influence over the 2016 Brexit Referendum and the US Presidential Election, while its absorption of teenage attention has led many to point out its negative effects on mental health and education.
We have not yet seen a significant row back on social media usage, but with AI, the potential for drastic measures is much larger. Suleyman believes there will be a moment with AI where we have to forego the benefits because of the risks.
The Great Opportunity: Abundance and Assisting the World’s Problems
Despite his cautious tone in The Coming Wave, Suleyman remains an AI optimist. When asked during the Q and A whether he thought AI would be a boon for bad actors or make us stupider (through having to learn less), Suleyman disagreed. He pointed out that there was a similar sentiment around bad actors and the early Internet, but that generally societies are strong enough to make the only fringe players.
In the case of education, Suleyman explained that within three years, everyone in the world will have a world class tutor in their pocket – thus the potential for human learning, as a collective species, is much higher. Because of this, AI will also be a massive enhancer of social mobility in the 21st century.
He was also very positive on AI’s potential to solve the world’s biggest problems. One of humanity’s greatest challenges remains removing carbon from the atmosphere through a cost efficient process. But AI may be able to solve the economic cost that hampers the ecological cleanup, and invent new materials that can help us become much more efficient.
Overall, Suleyman’s view was that AI will create abundance. If we think back to the first industrial revolution, the same happened. The population of industrialised countries boomed because of the surpluses created. We will probably not go through a similar population boom because of AI – instead, we will more likely receive an abundance of leisure time.
