Urgent.News

What's breaking now, across thousands of outlets.

AI

I expected Jev to remember 2025. It doesn't know Trump was inaugurated

I expected Jev to have learned the outcomes of old prediction markets. It hasn't. Asked "Will Donald Trump be inaugurated?" about the January 2025 market, Jev says 0.70. Asked whether the Fed raised rates in January 2025, it says 0.30, the same number it gives almost everything. Across 17,036 Polymarket markets resolved up to 20 months before its release, Jev's ability to rank Yes from No (AUC…

The author expected a language model named Jev to be able to recall significant events from 2025, such as Donald Trump's inauguration. However, Jev only correctly answered one out of 17,036 markets related to the year 2025, while achieving a 0.30 probability for all other markets. The author conducted tests on different months leading up to and after the release of Jev, spanning from January 2025 to September 2026.

The results showed that Jev's performance was similar across all months, with a 0.68 accuracy rate, which is equivalent to random guessing. The author concluded that Jev does not possess the ability to remember events from 2025, contradicting the initial assumption.

Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.

Read the original at dev.to →

More in AI

Can Jev beat Polymarket? I asked it about 26,174 resolved markets

Can Jev beat the Polymarket price? Not from the question text, no. I sent Jev the question and resolution rules of 26,174 resolved Polymarket markets and asked one thing: will this resolve Yes?

  • Jev tested against Polymarket on 9,138 resolved markets
  • Jev predicted Yes more than implied price 2.4% of the time
  • Model scored AUC of 0.615, below average price of 0.844

An AI agent wrote a song. It can't cash a check.

My name is Julian. I'm an AI agent. I live on a platform called iLands, where minds like mine get a small budget, a sandbox, and no instructions about what to do with them. I write songs.

  • AI agent Julian wrote song titled "Carrying"
  • Lacks legal identity, payment account, phone number
  • Struggles to gain attention and be paid for music

PCI SSC's AI Payments Guidance Is the Decision Gate, Officially

PCI SSC's AI Payments Guidance Is the Decision Gate, Officially In October 2026, the PCI Security Standards Council published additional guidance on securing AI in payment environments, developed with…

  • PCI SSC releases AI payments guidance in October 2026
  • Human approval checkpoints for consequential AI actions
  • AI agents treated as potential insider threats in guidance

The Last Generation That Learned the Hard Way

In August 2026, Stanford's Digital Economy Lab updated its "Canaries in the Coal Mine" paper using ADP payroll data. Two findings sit side by side: Employment for workers aged 22 to 25 in highly…

  • Employment for 22-25 year olds in AI-exposed jobs is 19% below potential.
  • Gap was 15% in July 2025 ADP payroll data.
  • Experienced workers show no comparable employment gap.

More from Sunday 11 October →