From 600+ conversations with the world’s leading thinkers.
If our future is to count as a utopia, we cannot allow a massive oppressed class of hyper-sentient, uncomfortable digital beings. We want it to be good for all kinds of minds.
Algorithms will soon surpass our innate ability to cater to our health, wellness, and overall well-being. As this unfolds, uncharted realms of purpose and significance will surface, ones we can't currently fathom. The act of humans making choices will become a mere memory. We'll be spectators in a novel arena.
As you think this through, a lot of structure dissolves. And then the question becomes: what remains once all of that is gone? You could still choose to do these activities, of course, but there would no longer be any point—no instrumental need. You would only do them simply because you wanted to…
At Axiom we've raised $64 million—we're a small startup—and we recently won the Putnam competition. We scored 90 out of 120, which would have placed us above all ~4,000 human contestants last year and at the level of a Putnam Fellow, meaning top five in the world. If you tried to achieve that purely through informal methods—where hallucination is a persistent risk—getting to the same level of consistent correctness would likely require a lot more resources.
These are technologies that are autonomous in many, many ways. They are independent in many, many ways – they have free will. They can replicate. And that makes a difference because then we teach them how to learn, but we have no idea what they will do with that ability to learn.
By the time he was still scrolling through his options, I simply showed him the answer on my phone. He was taken aback, admitting that this was a far superior approach. This experience exemplifies the revolutionary nature of answer engines.
I think the answer, really, in my view, is that we cannot control them because they are smarter. As simple as that, we only know that the smartest hacker in the room will always find a way through our defences. So maybe we should stop our arrogance for a minute.
Human cooperation is the greatest force in history- it enabled us to put people on the moon, build microprocessors and advance medicine. Human cooperation with intelligent machines will define the next era of history.
My fear has never been the machines waking up and deciding to do away with us, but rather that we- in our own bone headed way- deploy systems inappropriately, or without thinking through the unintended consequences that may occur.
While the cautionary tales narrated by these thinkers have merit, one can't help but feel they occasionally veer into hyperbole. They often highlight the idea that AI systems can self-evolve, enhancing their capabilities exponentially in mere seconds. However, this overlooks the fact that genuine intelligence augmentation necessitates the incorporation of vast new data.
A model that's really strong at mathematical reasoning is likely to be strong at coding. And a model that's excellent at both math and code is often very good at analysing the nuts and bolts of legal reasoning as well.
The ultimate goal is to be in a state of flow with machines. Think about people working with horses, or herding cattle with a dog, they are examples of interactions with other intelligent creatures in a way which is fluid and allows us to achieve something we couldn't do ourselves.