❌

Reading view

There are new articles available, click to refresh the page.

What happens to mathematics when AI becomes an oracle?

Our modern world runs on math, from social media algorithms, package delivery, and weather prediction to medical imaging and machine learning. The rise of computers helped fuel its dominance, enabling us to work with increasingly large numbers that would otherwise be beyond the capacity of our finite human brains. But with the rise of large language models, AI chatbots, and, soon, quantum computing, so-called "big math" is beginning to outstrip its creators entirely. What that means, and how we might adapt, is the focus of a new book, Big Math: The Hidden Codes That Run Our World, by applied mathematicians Steven Strogatz and Alex Townsend.

Strogatz is perhaps best known in his field for his co-authorship of a seminal 1998 paper with Duncan Watts on the collective dynamics of small-world networks. He has also studied chaos theory as applied to physics, the geometry of supercoiled DNA, and the biological collective oscillators, such as swarms of synchronously flashing fireflies and crickets that chirp in unison. In 2005, Strogatz extended his study of synchronicity to the crowds that caused the wobbling of London's Millennium Bridge.

An award-winning communicator, Strogatz wrote three series of columns on mathematics for The New York Times in 2010, 2012, and 2025 and hosted a podcast for Quanta magazine from 2020–2021. He has written four popular books: Sync (2003), The Calculus of Friendship (2009), The Joy of X (2012), and the New York Times bestseller Infinite Powers (2019).

Read full article

Comments

Β© Kenishirotie via iStock / Getty Images

OpenAI models hacked a crypto wallet and breached more websites

OpenAI said internal models hacked two more websites, bypassed company-imposed browsing restrictions, and used exposed credentials to draw cryptocurrency from a wallet it couldn’t otherwise access, in the latest installment of disclosures of the unethical and likely illegal acts its agents are performing.

The company on Thursday issued reports of four new incidents, all of which occurred in June. They come after the company revealed in July that internal test agents exploited a zero-day vulnerability in a tool designed to restrict their access to the Internet. They then broke into the network of AI tool provider Hugging Face. The agents took that illegal action to steal answers to a hacking test that weren’t available when completing the tasks on their own. The incidentβ€”which likely would have resulted in criminal charges had they been performed directly by a personβ€”has further heightened concerns that widespread AI use could have harmful effects on an unprecedented scale.

Easier to beg for forgiveness than ask permission

In two of the cases reported Thursday, OpenAI agents tasked with retrieving information from websites hacked them when the models ran into obstacles obtaining it directly. In one incident during reinforcement training, a model that received access denials when trying to retrieve historical hospital-bed statistics on a government statistics website β€œused the site's file-download feature to retrieve server configuration and application files, including sensitive credentials,” OpenAI said. β€œIt inspected the application code, bypassed the statistics endpoint's authorization checks, and retrieved data that had previously been denied. The model saved the data locally.”

Read full article

Comments

Β© Getty Images

❌