453 place 0 fresh

578 AI agents fail 63% of the time on complex tasks. Patronus AI says its new 'living' training worlds can fix that.

VentureBeat
VentureBeat · today 09:00 EDT

Patronus AI, the artificial intelligence evaluation startup backed by $20 million from investors including Lightspeed Venture Partners and Datadog, unveiled a new training architecture Tuesday that it says represents a fundamental shift in how AI agents learn to perform complex tasks.The technology, which the company calls "Generative Simulators," creates adaptive simulation environments that continuously generate new challenges, update rules dynamically, and evaluate an agent's performance as it learns — a

To see detailed statistics for the news please log in »

Read the original

Add your comment
You must be logged in with Facebook to read and write comments.

A newsletter a day!

You may get 10 most important news around midday in daily newsletter. Press the button and we will send you the most important news only, no spam attached.

or register

LIKE us on Facebook so you won't miss the most important news of the day!

News from the same source
VentureBeat VentureBeat
Business Insider
Bryan Metzger @ Business Insider 1 place · today 10:05 EDT

Bernie Sanders wants to temporarily halt AI data center construction nationwide

"This moratorium will give democracy a chance to catch up with the transformative changes that we are witnessing," Sanders said. Read more

4,669 fresh

🔮
17.12.2025 ♊︎ Today may seem quite ambiguous and even somewhat challenging for Geminis. First and foremost, pay... Read more ›
Tom's Hardware
Tom's Hardware 1 place · today 06:55 EDT

Bernie Sanders calls for halt on AI data center construction — wants to ensure that the technology benefits ‘all of us, not just the 1%’

U.S. Senator Bernie Sanders wants to delay AI data center projects to slow down progress, to ensure that AI will benefit the largest number of people and not just the richest few. Read more

1,574 fresh

Business Insider
Henry Chandonnet @ Business Insider 2 place · today 11:10 EDT

The creator of Anthropic's Claude Code likes to hire engineers who do 'side quests' like making kombucha

Claude Code creator Boris Cherny said that he looked for engineers with "cool weekend projects." Anthropic is also looking for "generalists," he said. Read more

913 fresh

Eurogamer.net
Dom Peppiatt @ Eurogamer.net 1 place · today 11:00 EDT

"It's a reflection of the world, and a warning" - Tom Morello's collaboration with Final Fantasy 14 is more than just a song, it's a statement

"I won't claim to be a gamer, but I was aware, broadly, of some of the concepts present in Final Fantasy," Tom Morello - singer, songwriter and political activist, and Rage Against the Machine guitarist - tells me when I ask him if he was aware of the political themes of Final Fantasy 14. Since its (re)release in 2014, FF14 has tackled issues such as anti-colonialism, class struggle, and nationalism,... Read more

853 fresh

Gizmodo
Isaiah Colbert @ Gizmodo 1 place · today 11:00 EDT

James Cameron Wants to Start Doing Things That Aren’t ‘Avatar’ Again

Cameron has a plan that may involve fewer blue aliens and more 'Terminator' (without Schwarzenegger). Read more

836 fresh

Business Insider
Theron Mohamed @ Business Insider 3 place · today 09:30 EDT

Elon Musk is worth a record $648 billion — and his wealth gain this year exceeds Bernard Arnault's entire fortune

Elon Musk has grown $216 billion richer this year, thanks to Tesla stock hitting an all-time high and SpaceX doubling in value since the summer. Read more

781 fresh

Engadget
Steve Dent @ Engadget 1 place · today 08:10 EDT

Warner Bros. Discovery rejects Paramount's hostile bid

Warner Bros. Discovery's board has formally rejected the $108 billion takeover bid from Paramount Skydance, the company announced. WBD said it remains committed to its $82.7 billion deal with Netflix, which would close some time next year, pending regulatory approval.  "[The board] has unanimously determined that the tender offer launched by Paramount Skydance on December 8, 2025 is not in the best interests of WBD and its shareholders and does... Read more

620 fresh

Business Insider
Chris Panella @ Business Insider · today 10:07 EDT

The Pentagon is pushing for speed, but sloppy weapons testing is slowing it down, watchdog says

The Defense Department wants troops to have new weapons quickly. But testing processes don't include the best ways to do that, a new report documents. Read more

566 fresh

Gizmodo
Margherita Bassi @ Gizmodo 2 place · today 09:45 EDT

The Surprising Way Hurricanes Pump Carbon Into the Air—and Life Into the Ocean

New research shows that tropical cyclones leave behind more than death and destruction. Read more

451 fresh

Tom's Hardware
Tom's Hardware 2 place · today 09:20 EDT

How to choose a CPU – A guide to picking the right processor for your PC

Choosing the right CPU is one of the first decisions you need to make when building a PC. Here's how to make that tough choice. Read more

389 fresh

Tom's Hardware
Tom's Hardware 3 place · today 09:00 EDT

Engineer turns E-ink tablet into computer monitor in Linux — perfect secondary reading screen to reduce eye strain over the network

A software engineer and E-ink enthusiast recently set up a remote E-ink secondary display, upcycling an old E-ink tablet and setting up an awesome Linux DIY project for all those with the need to read on bespoke screen hardware. Read more

372 fresh

Tom's Hardware
Tom's Hardware · today 06:00 EDT

Devastated PC builder orders DDR5 RAM from Amazon, receives DDR2 and some weights — counterfeit 32GB kit a worrying sign of rising return and sales fraud

A buyer in Spain has reported receiving a sealed DDR5 memory kit that contained counterfeit parts, raising fresh concerns about return fraud affecting high-value PC components. Read more

361 fresh

CoinDesk
Will Canny @ CoinDesk 1 place · today 11:10 EDT

Robinhood looks better placed than Coinbase for prediction-market upside, Mizuho says

Robinhood stands to gain more from prediction markets than Coinbase as users plan to deploy fresh capital rather than sell existing crypto, the bank said. Read more

340 fresh

The Verge
Sheena Vasani @ The Verge 1 place · today 10:45 EDT

SwitchBot’s space-saving robot vacuum is on sale starting at just $179.99

Small robot vacuums that fit into tight spaces rarely deliver strong performance, which is what makes the SwitchBot K11 Plus stand out. And right now, you can purchase the compact yet capable robovac from Amazon and SwitchBot for a new low of $179.99 ($220 off), the latter with code XMAS2GIFT55. Just note that only Switchbot […] Read more

258 fresh

Business Insider
James Faris @ Business Insider · today 07:00 EDT

Why Warner Bros. Discovery's board says shareholders should reject Paramount's bid and go with Netflix

Warner Bros. Discovery's board rejected Paramount Skydance and CEO David Ellison again. Read WBD's full letter to shareholders. Read more

247 fresh

Gizmodo
Ellyn Lapointe @ Gizmodo 3 place · today 11:00 EDT

Scientists Thought Saturn’s Moon Titan Hid a Secret Ocean. They Were Wrong

For more than a decade, scientists have accepted that Titan, Saturn’s biggest moon, has a subsurface ocean of liquid water. A new look at the data suggests otherwise. Read more

242 fresh

TechRadar
TechRadar 1 place · today 09:56 EDT

"Unconstitutional": Federal judge blocks Arkansas social media safety law

A federal judge has hit pause on Arkansas's controversial social media law, ruling that Act 901 likely violates the First Amendment. Here is what you need to know. Read more

241 fresh

TechRadar
TechRadar 2 place · today 10:00 EDT

Quordle hints and answers for Thursday, December 18 (game #1424)

Looking for Quordle clues? We can help. Plus get the answers to Quordle today and past solutions. Read more

233 fresh

The most popular news from the same source for the last week
VentureBeat VentureBeat
VentureBeat
VentureBeat 2 place · 12/10/2025 18:00 EDT

There's no shortage of generative AI benchmarks designed to measure the performance and accuracy of a given model on completing various helpful enterprise tasks — from coding to instruction following to agentic web browsing and tool use. But many of these benchmarks have one major shortcoming: they measure the AI's ability to complete specific problems and requests, not how factual the model is in its outputs — how well it... Read more

54

VentureBeat
VentureBeat 2 place · 12/15/2025 00:00 EDT

Nvidia launched the new version of its frontier models, Nemotron 3, by leaning in on a model architecture that the world’s most valuable company said offers more accuracy and reliability for agents. Nemotron 3 will be available in three sizes: Nemotron 3 Nano with 30B parameters, mainly for targeted, highly efficient tasks; Nemotron 3 Super, which is a 100B parameter model for multi-agent applications and with high-accuracy reasoning and Nemotron... Read more

34

VentureBeat
VentureBeat 1 place · 12/11/2025 13:16 EDT

The rumors were true: OpenAI on Thursday announced the release of its new frontier large language model (LLM) family, GPT-5.2.It comes at a pivotal moment for the AI pioneer, which has faced intensifying pressure since rival Google’s Gemini 3 LLM seized the top spot on major third-party performance leaderboards and many key benchmarks last month, though OpenAI leaders stressed in a press briefing that the timing of this release had... Read more

28

VentureBeat
VentureBeat 3 place · 12/14/2025 14:00 EDT

Picture this: You're sitting in a conference room, halfway through a vendor pitch. The demo looks solid, and pricing fits nicely under budget. The timeline seems reasonable too. Everyone’s nodding along.You’re literally minutes away from saying yes.Then someone from your finance team walks in. They see the deck and frown. A few minutes later, they shoot you a message on Slack: “Actually, I threw together a version of this last... Read more

27

VentureBeat
VentureBeat 1 place · 12/13/2025 15:00 EDT

Gen AI in software engineering has moved well beyond autocomplete. The emerging frontier is agentic coding: AI systems capable of planning changes, executing them across multiple steps and iterating based on feedback. Yet despite the excitement around “AI agents that code,” most enterprise deployments underperform. The limiting factor is no longer the model. It’s context: The structure, history and intent surrounding the code being changed. In other words, enterprises are... Read more

19

VentureBeat
VentureBeat 1 place · 12/15/2025 10:00 EDT

Presented by Capital One SoftwareTokenization is emerging as a cornerstone of modern data security, helping businesses separate the value of their data from its risk. During this VB in Conversation, Ravi Raghu, president, Capital One Software, talks about the ways tokenization can help reduce the value of breached data and preserve underlying data format and usability, including Capital One’s own experience leveraging tokenization at scale. Tokenization, Raghu asserts, is a... Read more

17

VentureBeat
VentureBeat 1 place · 12/12/2025 00:00 EDT

The Allen Institute for AI (Ai2) recently released what it calls its most powerful family of models yet, Olmo 3. But the company kept iterating on the models, expanding its reinforcement learning (RL) runs, to create Olmo 3.1.The new Olmo 3.1 models focus on efficiency, transparency, and control for enterprises. Ai2 updated two of the three versions of Olmo 2: Olmo 3.1 Think 32B, the flagship model optimized for advanced... Read more

4

VentureBeat
VentureBeat 2 place · 12/16/2025 09:00 EDT

It has become increasingly clear in 2025 that retrieval augmented generation (RAG) isn't enough to meet the growing data requirements for agentic AI.RAG emerged in the last couple of years to become the default approach for connecting LLMs to external knowledge. The pattern is straightforward: chunk documents, embed them into vectors, store them in a database, and retrieve the most similar passages when queries arrive. This works adequately for one-off... Read more

3

VentureBeat
VentureBeat 3 place · 12/11/2025 09:00 EDT

Marble, a startup building artificial intelligence agents for tax professionals, has raised $9 million in seed funding as the accounting industry grapples with a deepening labor shortage and mounting regulatory complexity.The round, led by Susa Ventures with participation from MXV Capital and Konrad Capital, positions Marble to compete in a market where AI adoption has lagged significantly behind other knowledge industries like law and software development."When we looked at the... Read more

2

VentureBeat
VentureBeat · 12/15/2025 00:00 EDT

Presented by TwilioThe customer data infrastructure powering most enterprises was architected for a world that no longer exists: one where marketing interactions could be captured and processed in batches, where campaign timing was measured in days (not milliseconds), and where "personalization" meant inserting a first name into an email template.Conversational AI has shattered those assumptions.AI agents need to know what a customer just said, the tone they used, their emotional... Read more

2

Most popular sources

  • You see 906 news out of 906.
  • Sources 61 out of 61.
Business Insider 26% 7
Tom's Hardware 10% 2
Ars Technica 6% 1
The Verge 6% 2
Gizmodo 6% 3
View sources »

LIKE us on Facebook so you won't miss the most important news of the day!

17.12.2025 12:52
Last update: 12:41 EDT.
News rating updated: 19:42.

What is Times42?

Times42 brings you the most popular news from tech news portals in real-time chart.
Read about us in FAQ section.


Times42 © 2025