Raymond J. de Souza: Restraining AI is a task for natural intelligence

Anthropic CEO Dario Amodei is among the big tech leaders calling for AI researchers to apply the brakes to ensure the safety of mankind. His comments followed the resignation of Anthropic researcher Jabob Coxan, who warned that “The people building AI earnestly believe that it could kill us all by the end of the decade." Photo by FABRICE COFFRINI / AFP via Getty Images

Article content

“If this technology is coming, it must go well,” said Christopher Olah, founder of AI giant Anthropic, last May at the Vatican, speaking alongside Pope Leo XIV.

Article content

On Monday, Bell Canada announced a massive $50-billion AI investment in Saskatchewan. The announcement included Prime Minister Mark Carney, as well as his predecessors Jean Chrétien and Stephen Harper. You might otherwise need a papal funeral to get three PMs on hand. That’s how big AI now is.

Article content

Article content

The technology is not only coming. It is here. Is it going well?

Article content

Last week Jacob Coxon resigned from Anthropic, warning on X that neither OpenAI nor Anthropic “is acting responsibly” but are “gambling with our lives.”

Article content

Article content

Article content

“The people building AI earnestly believe that it could kill us all by the end of the decade,” Coxon added, setting off a global debate that had some political leaders, previously AI enthusiasts, running to revise their positions. There is a growing popular movement against AI data centres and their rapacious appetite for land, for energy, for raw materials, for microchips, for debt.

Article content

Article content

On Saturday, Anthropic CEO Dario Amodei called for a slowing down of AI development — “pacing” — so that it might be possible for AI advances to become a “race to the top” as opposed to the current “race to the bottom.”

Article content

“If slowing down bought us even an extra year or two before models reach critical levels of capability, and we used that time to advance alignment, we could greatly reduce the risk that something goes seriously wrong,” Amodei wrote.

Article content

President Donald Trump, for this part, mocked Amodei, saying that the only guardrails AI needs are a “STRONG AND SMART (High IQ!) PRESIDENT.” Rest easy, then.

Article content

It is refreshingly remarkable that the titans of Big Tech are issuing stark warnings. These have not been trustworthy people when it comes to mitigating the maleffects of their work. Quite the opposite.

Article content

Article content

The late Steve Jobs was a generational force in imagination, engineering and design. But he did more than anyone else to addict billions — including children — to their screens. He knew the danger, which is why he did not allow his own children to use the iPad. Apple’s marketing did not include such advice for other parents.

Article content

Article content

Just last month, Mark Zuckerberg’s Meta agreed to a settlement — a whopping US$18 billion, but not a significant dent in Meta’s budget — that included modifying its algorithms to make them less addictive to minors. Which meant that their addiction-promoting algorithms were a choice that they had previously made, with malice-for-money aforethought.

Article content

An enormous number of highly talented folks in the tech sector devote their energies to getting more and more people to spend more and more of their time and money on gaming, gossip, gambling and pornography. The sector, by and large, is not noted for restraint in promoting distracting and destructive things.

Article content

So when the tech barons say that it is time to reflect, time to worry, time to pull back, time not to do something just because we can do it — then it is high time to listen. Pace, pause, put a stop to AI for AI’s sake. Make the priority AI for humanity’s sake.