AI Summit: Time to Shift Focus from Doomsday Scenarios to Present Threats

JJohn October 29, 2023 1:02 PM

The AI safety summit this week will see a shift in focus, as experts argue against the fixation on doomsday scenarios. Instead, the immediate threat of AI-induced misinformation needs immediate attention, warns Aidan Gomez, a leading industry figure and co-founder of Cohere.

Misinformation: An immediate threat from AI

Aidan Gomez is a well-respected figure in the world of artificial intelligence, having co-authored a research paper that played a pivotal role in the creation of chatbot technology. Gomez argues that while the industry should continue to investigate potential long-term existential threats posed by AI, the focus of public policy should be on more immediate, tangible risks. He singles out the large-scale generation of misinformation as a key concern, noting that the preoccupation with doomsday scenarios can often serve as a distraction from such immediate harms.

AI-powered tools, including chatbots and image generators, are no longer future inventions; they're here and in widespread use. Such tools, powered by advanced large language models, are astonishing the public with their abilities to create realistic text and images from mere text prompts. These technologies are already integrated into billion-user products, increasing the urgency to address associated risks, which, while not existential, can have immediate and significant impacts on society.

AI-created media: A new challenge in misinformation

The ability of AI models to generate convincingly human-like content is a boon for many applications, but it poses a significant threat in the realm of misinformation. Gomez singles out this issue as his key concern. AI-generated content can be extremely compelling, often indistinguishable from human-created media. This presents a formidable challenge to society, necessitating urgent measures to equip the public with the ability to distinguish between AI-generated and human-created content.

Addressing AI risks: Summit discussions on misinformation and social trust

The AI safety summit will feature an extensive array of discussions revolving around AI and its implications. One of the key areas of focus will be the role of AI in misinformation and its potential to disrupt elections and erode social trust. These sessions aim to shed light on the pressing need for concrete steps to mitigate AI risks and to foster a better understanding of the immediate threats posed by AI.

While certain tech professionals and experts have expressed concerns about the existential threat posed by AI, many others, like Yann LeCun, dismiss such fears as unfounded. LeCun, the chief AI scientist at Meta and a co-winner of the highly prestigious ACM Turing award, opines that AI reaching human-level intelligence—a point where a system could potentially evade human control—is currently far from reality. He adds that intelligence doesn't equate to the desire to dominate, debunking the notion that AI might wipe out humanity as 'preposterous'.

More articles

Also read

Here are some interesting articles on other sites from our network.