Shuby headshot

DeepMind’s CEO says using AI can make you a genius — or hurt your critical thinking skills

by with No Comment Xpandfy

It’s up to you whether AI makes you sharper or slowly dulls your brain, says Demis Hassabis.

In a Thursday interview with entrepreneur Varun Mayya on the sidelines of the India AI Impact Summit, the Google DeepMind CEO said that AI is just like the internet. People can use it to learn all kinds of topics, or use it in ways that “degrade” their thinking.

“With AI, if you use it in a lazy way, it will make you worse at critical thinking and so on,” he said. “But that’s down to you as the individual. No one can help you do that.”

He added that people need to be smart and use these technologies in ways that enhance their thinking rather than dull it.

Hassabis cofounded DeepMind in 2010, which Google acquired in 2014. It merged with Google Brain in 2023 to form Google DeepMind, the lab behind tools such as Gemini and Nano Banana. The CEO and a DeepMind coworker, John Jumper, were awarded the 2024 Nobel Prize for Chemistry for their work on protein structure prediction.

As AI gets incorporated into daily life, debates about its risks and rewards have intensified, with several tech leaders warning about the dangers of an overreliance on AI tools.

Earlier this week, tech billionaire Mark Cuban said that there are two types of people who use AI.

“There are generally 2 types of LLM users, those that use it to learn everything, and those that use it so they don’t have to learn anything,” Cuban said of large language models in an X post on Tuesday.

Cuban has previously said that AI models can’t provide all the answers and are “stupid” but like “a savant that remembers everything.”

At a June conference, the CEO of French AI lab Mistral said that a risk of using AI for everything is that humans will stop trying.

“The biggest risk with AI is not that it will outsmart us or become uncontrollable, but that it will make us too comfortable, too dependent, and ultimately too lazy to think or act for ourselves,” Arthur Mensch said.




Source link

Aditi Bharade

DeepMind’s CEO said there are still 3 areas where AGI systems can’t match real intelligence

by with No Comment Xpandfy

True artificial general intelligence is on the way, but it still has some ways to go, said Google DeepMind’s CEO.

Speaking at an AI summit in New Delhi, Demis Hassabis was asked whether current AGI systems can match human intelligence. AGI is a hypothetical form of machine intelligence that can reason like people and solve problems using methods it was not trained in.

Hassabis’ short answer: “I don’t think we are there yet.”

He listed three areas where current AGI systems are falling short. The first was what he called “continual learning,” saying that the systems are frozen based on the training they received before implementation.

“What you’d like is for those systems to continually learn online from experience, to learn from the context they’re in, maybe personalize to the situation and the tasks that you have for them,” he said during the discussion.

Secondly, Hassabis said current systems struggle with long-term thinking.

“They can plan over the short term, but over the longer term, the way that we can plan over years, they don’t really have that capability at the moment,” he said.

And lastly, he said that the systems lack consistency. They’re adept in some areas and unskilled in others.

“So, for example, today’s systems can get gold medals in the international Math Olympiad, really hard problems, but sometimes can still make mistakes on elementary maths if you pose the question in a certain way,” he said. “A true general intelligence system shouldn’t have that kind of jaggedness.”

Humans, in comparison, would not make mistakes on an easy math problem if they were math experts, he added.

Hassabis said in a “60 Minutes” interview last year that true AGI would arrive in five to 10 years.

The executive cofounded DeepMind, an AI research lab, in 2010. The lab was acquired by Google in 2014 and is the brains behind Google’s Gemini. In 2024, Hassabis won a joint Nobel Prize in chemistry for his work on protein structure prediction.

AGI is a disputed topic in Silicon Valley. Databricks CEO Ali Ghodsi said at a September conference that current AI chatbots already meet the definition of AGI, but Silicon Valley leaders keep “moving the goalposts” and pushing toward superintelligence, or AI that can outthink humans.

The AI Summit in India, from Monday to Friday this week, has attracted big names from the tech and AI spheres. Notable speakers on the summit’s agenda include OpenAI CEO Sam Altman, Anthropic CEO Dario Amodei, Google CEO Sundar Pichai, and Meta’s chief AI officer, Alexandr Wang.




Source link

Shuby headshot

DeepMind’s CEO does a ‘second day’ of work at night: ‘I come alive at about 1 a.m.’

by with No Comment Xpandfy

You’re not the only one with a strange sleep routine.

In a video interview with Fortune released on Wednesday, Google DeepMind CEO Demis Hassabis said he sleeps very little and splits his waking hours into two working days.

“I do try and get six, but I have unusual sleeping habits,” he said, about his number of hours of sleep. “I sort of manage during the day.”

He said that any less sleep than that would be bad for the brain.

Hassabis cofounded DeepMind in 2010, which Google acquired in 2014. It merged with Google Brain in 2023 to form Google DeepMind, the lab behind tools such as Gemini and Nano Banana. The CEO and a DeepMind coworker, John Jumper, were awarded the 2024 Nobel Prize for Chemistry for their work on protein structure prediction.

Hassabis said that he tries to pack his day in the office with as many meetings as possible with “almost no time” in between. He then gets home, spends time with family, and has dinner.

“Then I sort of start a second day of work about 10 p.m. and go to 4 a.m., where I do my thinking and kind of more creative work and research work,” he said.

The CEO added that he’s followed this schedule for about a decade. Hassabis earlier spoke about his sleep routine in a 2017 interview with BBC Radio.

“I can’t imagine being creative at four in the morning. But, I come alive at about 1 a.m.,” he told Fortune’s Alyson Shontell.

Hassabis’ routine matches what other tech founders have shared about their sleep schedules, especially during the early stages of founding or growing their businesses.

Elon Musk has said that he functions best with about six hours of sleep — any less affects his performance. In a 2018 interview with Bloomberg, Musk said that he slept on the floor of a Tesla factory during some production periods.

Marc Benioff, meanwhile, said in a 2023 interview that he averages about eight hours of sleep.




Source link