Connect with us

Software Technology

AI pioneer Geoffrey Hinton rings the alarm on the potential dangers of advanced AI as he resigns from Google.



Geoffrey Hinton
Geoffrey Hinton

Experts including Elon Musk urge caution as AI advances too fast – say “We need to step back

Geoffrey Hinton, an eminent figure in the field of artificial intelligence and widely recognized as the godfather of AI, has resigned from Google and voiced his concern about the increasing dangers stemming from advancements in the industry.

Hinton expressed sorrow for his prior work in a statement to the New York Times and emphasised the urgent need for more ethical approaches to AI. He also cautioned about the dangers that could come with AI chatbots, saying that they might soon surpass humans in intelligence.

“Right now, they’re not more intelligent than us, as far as I can tell. But I think they soon may be.”

Hinton underlined his worries and encouraged the industry to give ethical considerations first priority while creating AI technology in an interview with the BBC. He emphasised the significance of being proactive rather than reactive in identifying and addressing potential dangers.

We must carefully evaluate the consequences of our actions as the area of artificial intelligence develops and expands, and we must cooperate to build a future in which AI helps society as a whole.

Dr. Hinton is a cognitive psychologist and computer scientist of British and Canadian descent. He has pioneered ground-breaking work in the fields of neural networks and deep learning, providing the groundwork for the cutting-edge AI systems we now use, such as ChatGPT.

Dr. Hinton is also a leading figure in the development of neural networks that are based on the capacity of the human brain to learn and process information. Deep learning is a cutting-edge technology that has completely changed the area of artificial intelligence by allowing machines to learn and develop over time through experience, just like people.

In a BBC interview he admitted that he decided to leave the computer giant because of his advanced age. He decided it was time to take a well-earned retirement at the age of 75 and leave the fast-paced world of technology behind.

The chatGPT era

But Dr. Hinton’s influence on the field of artificial intelligence will undoubtedly last for years to come. He thinks that AI chatbots, like ChatGPT, are about to outperform humans in terms of information processing, which will change how we engage with technology and one another. Dr. Hinton will leave behind a lasting legacy as the inventor of deep learning, influencing the development of AI and its social effect.

“Right now, what we’re seeing is things like GPT-4 eclipses a person in the amount of general knowledge it has and it eclipses them by a long way. In terms of reasoning, it’s not as good, but it does already do simple reasoning. And given the rate of progress, we expect things to get better quite fast. So we need to worry about that.” he said.

What happens if AI is exploited

In an article featured in The New York Times, Dr. Hinton expressed his concerns about the potential for “malicious entities” to exploit AI technology for their own nefarious purposes.

When asked to expand on this issue during a recent interview with the BBC, he explained that it was essentially a worst-case scenario, a nightmare scenario that could come to fruition.

Dr. Hinton proceeded to offer an example of how this type of malevolent AI exploitation might play out, saying that a “bad actor” such as Russian President Vladimir Putin could hypothetically grant robots the ability to create their own subgoals.

However, Dr. Hinton cautioned that such a decision could ultimately lead to the creation of subgoals that revolve around obtaining more power, potentially paving the way for disastrous consequences.

AI vs Human Intelligence

Dr. Geoffrey Hinton discussed his findings of the key distinctions between biological intelligence and artificial intelligence (AI) in a BBC interview. His remarks have received a lot of attention due to his prominence in the field of artificial intelligence, and some have interpreted them as a warning about the possible risks this technology may present.

Mr. Hinton asserts that the intelligence we are gaining through AI differs significantly from the intelligence we already have as biological systems. Digital systems are created using sets of weights and models of the environment, whereas we rely on sophisticated neural networks to analyse information and make judgements. These Digital systems are not constrained by the amount of neurons and connections in our brains, as is the case with biological systems.

Digital systems may produce numerous copies of the same set of weights or model, allowing them to learn independently while quickly sharing their knowledge, in contrast to biological systems, which are constrained by the amount of neurons and connections in our brains.

One of the main benefits of AI systems like chatbots, which can access enormous amounts of information and learn at an unparalleled rate, is their capacity to share knowledge. It’s like having 10,000 people working on the same issue, each one adding their expertise and insights to the group, says Dr. Hinton. Because of this, AI systems can now learn much more than a single person ever could, potentially revolutionising industries like healthcare, banking, and transportation.

Following Dr. Hinton’s announcement, Matt Clifford, the Chairman of the UK’s Advanced Research and Invention Agency, emphasises the urgent need for investment in AI safety and control. He acknowledges the immense promise of this technology, but he issues a warning that it is crucial to reduce any hazards and make sure that AI is developed in a way that benefits society.

We must step back for a while

In an open letter published in March and co-signed by numerous experts in the field of artificial intelligence, including tech billionaire Elon Musk, a request for a halt to all advancements more advanced than the present version of AI chatbot ChatGPT was made. This was done to allow for the creation and implementation of strong safety measures. Warren Buffet in a recent interview showed concerns about the accelerating growth of AI and their potential to reduce employment in certain fields.

Yoshua Bengio, another so-called “godfather of AI,” who shared the 2018 Turing Award for deep learning research with Dr. Hinton and Yann LeCun, also signed the letter.

Mr. Bengio said that “we need to take a step back” due to the “unexpected acceleration” in AI systems.

The responsible move

Dr. Hinton emphasised that the tech giant had been “very responsible” and that he did not wish to attack Google

“I actually have some positive things to say about Google. And if I don’t work for Google, they have greater credibility”

Google’s top scientist Jeff Dean stated in a statement: “We continue to be dedicated to a responsible approach to AI. We are always learning to innovate bravely while also understanding growing threats.”

However, Dr. Hinton told the BBC that “in the shorter term” he believed that risks from AI would be greatly outweighed by benefits. “So I don’t think we should stop developing this stuff,” he added.

He added that a break would be challenging due to worldwide competitiveness. “Even if everybody in the US stopped developing it, China would just get a big lead,” he claimed.

The government must make sure AI is built “with a lot of thought into how to stop it going rogue,” according to Dr. Hinton, who also asserted that he is an expert on science, not policy.

AI is not just about Chatbots

Even though they are the most well-known at the moment, AI chatbots are only one type of artificial intelligence.

The algorithms that determine what videos you should next view on video streaming services are powered by AI. It can be used to screen job applications, to determine insurance rates, to diagnose medical issues (although human doctors still have the final word), and in recruitment.

The development of AGI, or artificial general intelligence, which can be taught to perform a variety of tasks within a scope, is what we are currently witnessing. As an illustration, ChatGPT can only provide text responses to a query, but as we are seeing, there are countless options inside that.

But even its designers were taken aback by the speed at which AI developed. Since Dr. Hinton created a ground-breaking image analysis neural network in 2012, it has radically changed.

Even Google CEO Sundar Pichai admitted in a recent interview that he was still learning about the capabilities of Bard, the company’s AI chatbot.

There is no doubt that we are travelling at high speeds, and some worry that the train may eventually begin constructing its own rails.


Copyright © 2023 Futurfeed