Head of Google's AI explains when we're going to see human-level AI but issues worrying warning
Artificial intelligence gaining a level of intelligence comparable to humans is generally the point in the science fiction movie where it becomes self-aware and tries to wipe us all out.
In recent years, AI has come on in leaps and bounds. There are some incredible things it could do to help us and there are some terrible impacts it could have if we're not careful about what we do with it.
Some of the latest AI video developments are scarily close to looking realistic now, especially when you consider where they were just a couple of years ago, and the future of this technology is likely to further worry those who are already concerned.
An expert voice on the topic of AI development is Demis Hassabis, the CEO of Google's DeepMind AI, who has given his prediction on when he thinks artificial general intelligence (AGI) will begin to rival humans in levels of intelligence.
Advert
You see, while there's a lot AI can do, it has trouble making much sense so its ability to generate something isn't always matched by an ability to actually understand it.
Researchers have found that the stuff AI can put out is 'fluent but incoherent', but once you dig down into it things don't actually make sense.
However, the day when some thinking AIs could rival human intelligence is apparently not that far off according to Hassabis, as CNBC reported that he warned some AI systems could get as smart as humans within the next five to 10 years.
He said: "I think today’s systems, they’re very passive, but there’s still a lot of things they can’t do.
"But I think over the next five to 10 years, a lot of those capabilities will start coming to the fore and we’ll start moving towards what we call artificial general intelligence."
He described an AGI as 'a system that’s able to exhibit all the complicated capabilities that humans can', and said that while we're not there yet, it's something we'll have to contend with soon enough.
The predictions made by the Google AI boss are actually less worrisome than those given out by several of his peers.
CNBC reported that Dario Amodei, CEO of AI start-up Anthropic, told them he reckoned it'd only be 'two or three years' before companies developed an AI that was 'better than almost all humans at almost all tasks.'
Hassabis gave a warning to the Hard Fork podcast that people needed to be ready for AI to get smarter, and the consequences of that.
"Whatever happens with these AI tools, you’ll be better off understanding how they work, and how they function, and what you can do with them," he said.
He recommended that students 'immerse yourself' in new technologies so they could be ready for the impact this would have on the world of work.
It seems AI is here to stay, whether we like it or not.
Topics: AI, Technology, Artificial Intelligence, Google
Joe Harker