3 place 0

315 OpenAI Announces Benchmarks for AI Life Sciences Research. Its Best Model Failed 63.9% of the Test

Slashdot
EditorDavid @ Slashdot · 06/20/2026 17:34 EDT

OpenAI Announces Benchmarks for AI Life Sciences Research. Its Best Model Failed 63.9% of the Test

This week OpenAI announced a 750-task test to to measure "whether AI systems can support realistic life science research tasks, not just answer biology questions."

But while OpenAI's top-performing GPT-Rosalind model led the rankings, Slashdot reader BrianFagioli notes that "it achieved a pass rate of just 36.1 percent, failing nearly two-thirds of benchmark tasks." Nerds.xyz points out that means "the best-performing model failed nearly two-thirds of the benchmark's tasks."




The benchmark also reveale

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
Slashdot Slashdot
Silicon Valley
George Avalos @ Silicon Valley 1 place · 02/07/2106 01:28 EDT

Newark apartment complex bought for much less than prior value

An East Bay apartment complex has been bought at a price that's well below its prior value. Read more

0

🔮
10.07.2026 ♎︎ Dear Libras, today will be a busy day filled with various events and emotions. Try... Read more ›
Silicon Valley
George Avalos @ Silicon Valley 2 place · 02/07/2106 01:28 EDT

PG&E buys San Jose building to bolster South Bay operations

A PG&E Corp. unit has bought a San Jose building in a move to bolster the utility's South Bay operations. Read more

0

SlashGear
SlashGear 1 place · today 21:45 EDT

Why Do Some Speed Limit Signs Have Fractions?

Speed limit signs are typically pretty standard, but every once in a while, you'll encounter one with a fraction on it. Here's why that happens. Read more

0 fresh

Engadget
Engadget 1 place · today 21:33 EDT

Meta deactivates feature that let you generate AI images of any public Instagram account

Meta has deactivated the Muse Image capability to create AI deepfakes of any public Instagram account you @-mention. Read more

0 fresh

GSMArena.com
GSMArena.com 1 place · today 21:23 EDT

Tecno Camon 50 Ultra's India launch date announced

The Tecno Camon 50 Ultra, unveiled in March, will make its debut in India on July 17. Tecno is advertising the Camon 50 Ultra in India as a smartphone "designed for creators, storytellers, and photography enthusiasts." The brand also says that Camon 50 Ultra is "built to capture life's best moments as they happen." Tecno hasn't revealed the Camon 50 Ultra's pricing and availability for India yet, so we'll have... Read more

0 fresh

Silicon Canals
Silicon Canals Editorial Team @ Silicon Canals 1 place · today 21:19 EDT

The quiet grief of outgrowing a friendship neither of you did anything to break

Some friendships don't end in conflict — they simply thin until neither of you is quite sure when the last real conversation happened. Psychologists call this ambiguous loss, and it may be one of the most under-grieved experiences of adult life. Read more

0 fresh

Wired
Maxwell Zeff @ Wired 1 place · today 21:07 EDT

OpenAI’s Head of Safety Is Leaving the Company

Johannes Heidecke’s departure comes as OpenAI tries to further integrate its research and safety teams. Read more

0 fresh

Silicon Canals
Silicon Canals Editorial Team @ Silicon Canals 2 place · today 21:04 EDT

The Boring Billion is the name geologists give to the stretch between roughly 1.8 and 0.8 billion years ago, when low-oxygen oceans slowed complex life — but evolution did not pause and Earth was not simply a billion years of slime

For roughly a billion years — from 1.8 to 0.8 billion years ago — Earth's oceans went chemically stagnant, oxygen stalled, and life did nothing much except reproduce as microbial slime. Geologists call it the Boring Billion, and it is the longest uneventful chapter in the planet's history. Read more

0 fresh

Digital Trends
Sudhanshu Kumar Mangalam @ Digital Trends 1 place · today 20:56 EDT

YouTube’s AI-powered search is rolling out in the US to find videos based on situations you describe

US users can now ask YouTube conversational questions to find videos for specific situations, ideas, and activities. Read more

0 fresh

Business Insider
Katherine Tangalakis-Lippert,Henry Chandonnet @ Business Insider 1 place · today 20:55 EDT

AI-enabled cheating is forcing some schools to go analog

University of Chicago Law School has announced a laptop ban to foster independent thinking amid AI's rise in legal education and practice. Read more

0 fresh

Startups News
Daniel Levi @ Startups News 1 place · today 20:32 EDT

GPT-5.6 vs. Claude Fable 5: Which AI Model Is Better? The Ultimate 2026 Comparison

OpenAI and Anthropic are once again competing for the top spot in enterprise AI, with both companies unveiling their most advanced models within weeks of each other. On June 10, Anthropic introduced Claude Fable 5, bringing its once-restricted Mythos-class AI ... Read more

0 fresh

The Verge
Jay Peters @ The Verge 1 place · today 20:10 EDT

No, Flock isn’t threatening people for debating surveillance

On Thursday, the Instagram account for a lecture series in Newport Beach, CA posted a photo of what appeared to be a cease and desist letter from the surveillance technology company Flock Safety. Flock has received significant backlash over its technology and work with law enforcement agencies, and this letter kicked off yet another wave […] Read more

0 fresh

Silicon Canals
Silicon Canals Editorial Team @ Silicon Canals · today 20:00 EDT

Amazon now runs more than a million warehouse robots — closing in on its roughly 1.2 million-strong operations workforce

In June of 2025, Amazon said it had put its one millionth robot to work, a unit sent to a warehouse in Japan, across more than 300 sites worldwide. “We’ve just deployed our 1 millionth robot, building on our position as the world’s largest manufacturer and operator of mobile robotics,” said Scott Dresser, a vice ... Read more Read more

0 fresh

The Verge
Jay Peters @ The Verge 2 place · today 19:49 EDT

Meta turns off the Instagram feature that users make AI deepfakes of public accounts

Following significant backlash, Meta is turning off the feature it announced this week that let users generate AI images based on content from public Instagram accounts just by tagging them. The feature, as originally set up, meant that content from any public Instagram account could be used in AI creations without the account owner's permission. […] Read more

0 fresh

SlashGear
SlashGear 2 place · today 19:45 EDT

This One Honda Is Selling Better Than All Mazdas Combined

Mazda builds incredible enthusiast vehicles, but one single crossover is currently outselling its entire North American lineup. Discover which one. Read more

0 fresh

Digital Trends
Shikhar Mehrotra @ Digital Trends 2 place · today 19:35 EDT

Meta just pulled its most controversial AI image generation feature days after launch

Meta launched then pulled a feature letting anyone generate AI images of public Instagram users who were opted in by default. Read more

0 fresh

The most popular news from the same source for the last week
Slashdot Slashdot
Slashdot
BeauHD @ Slashdot · 07/03/2026 22:00 EDT

Alibaba To Ban Claude Code In Workplace Over Alleged Backdoor Risks

Alibaba has reportedly banned employees from using Anthropic's Claude Code and directed them to its own Qoder platform amid a growing dispute over features that can help identify China-linked users. Reuters reports: The ban is part of a deepening spat between the two companies after Anthropic accused Alibaba of illicitly extracting its Claude AI model capabilities -- a dispute that highlights the frantic race between the U.S. and China to... Read more

0

Slashdot
BeauHD @ Slashdot · 07/04/2026 03:00 EDT

Video Game History Foundation Says Piracy Remains the Only Viable Preservation Method

An anonymous reader quotes a report from TechSpot: Video Game History Foundation founder Frank Cifaldi recently supported claims that piracy is the only effective way to preserve video games. The comments lay the blame squarely on game companies' refusal to keep legacy content available or allow archivists to build legal repositories. Sony's announcement that all PlayStation games will be digital-only from 2028 onward has sparked concern that titles will become... Read more

0

Slashdot
BeauHD @ Slashdot · 07/04/2026 07:00 EDT

Startup Targets Datacenters With 3D-Printed Nuclear Reactor Module

Startup Ampera has unveiled what it calls the first 3D-printed nuclear reactor module, built around a silicon-carbide core and pressure vessel designed for a thorium-based microreactor. The company says future systems could deliver 15 or 30 megawatts for up to 30 years without refueling. When The Register asked about availability, their spokesperson said: "We expect the power generation portion of the system to be available as early as 2027, with... Read more

0

Slashdot
EditorDavid @ Slashdot · 07/04/2026 10:34 EDT

What Is a Quantum Computer Good For? Absolutely Nothing - Yet

The Verge argues that researchers "have made genuine progress in quantum computing — it's just been largely incremental and too esoteric to immediately capture the public's imagination." And there are predictions that quantum computers will finally do something useful as soon as 2028: The drama can overshadow the real progress in quantum computing... Researchers have improved the qubits themselves, so they hold onto information longer. When they hold onto information... Read more

0

Slashdot
EditorDavid @ Slashdot · 07/04/2026 11:34 EDT

Decades-Old Bash Tricks Expose AI Coding Agents To Supply Chain Attacks

Slashdot reader wiredmikey writes: AI security researchers have uncovered a structural security flaw dubbed GuardFall that allows decades-old Bash shell tricks to bypass safeguards in most open source AI coding agents. By exploiting shell behaviors such as quote removal and variable expansion, attackers can hide malicious commands in repositories, README files, Makefiles, or other content consumed by AI agents. If executed — particularly in auto-approve or CI environments—the commands can... Read more

0

Slashdot
EditorDavid @ Slashdot · 07/04/2026 12:34 EDT

EchoStar's US Satellite Pay-TV Provider Dish DBS Files for Bankruptcy

EchoStar's satellite pay-TV unit Dish DBS has filed for Chapter 11 bankruptcy protection, reports Reuters. The move also applies to its wireless subsidiaries, according to the article, and "facilitates the wind-down of Dish Wireless's 5G network operations following an unexpected delay in a spectrum license sale to AT&T... under which EchoStar agreed to sell about 50 megahertz of its nationwide spectrum for $23 billion." Some context from Deadline.com: Charlie Ergen,... Read more

0

Slashdot
EditorDavid @ Slashdot · 07/04/2026 13:34 EDT

AOL's Owner Bending Spoons Hits Wall Street with $1.7 billion IPO

"The owner of AOL and other tech businesses hit Wall Street with a $1.7 billion initial public offering Wednesday," reports the Associated Press: The company is getting $1 billion in proceeds, while the rest is going to shareholders. The stock surged 39.7% in its first day of trading under the symbol "BSP" on the Nasdaq, giving it a market value of $25.2 billion. Among the company's well-known holdings are the... Read more

0

Slashdot
EditorDavid @ Slashdot · 07/04/2026 14:34 EDT

FSF Shares Update on 'LibrePhone' and New Automated Site Monitoring Tool

At the end of 2025, the FSF launched LibrePhone project, which is working to "better understand and reverse-engineer the nonfree blobs used by a great majority of (if not all) system on a chip designs available today." The FSF's summer newsletter shares this update: We started with researching the proprietary files in Android phones supported by the Lineage project, an Android-based volunteer-led mobile phone operating system with much free software... Read more

0

Slashdot
EditorDavid @ Slashdot · 07/04/2026 15:34 EDT

Did Microsoft Shift Its Profits to Low-Tax Countries?

Microsoft is apparently shifting its profits to countries with low taxes — and out of countries where they have many more employees and significant sales. Back in 2005 Former Microsoft CEO Steve Ballmer even said that a low corporate tax rate "is part of the overall advantage of doing business in Ireland," remembers long-time Slashdot reader theodp. (Ballmer added "It would be disingenuous to say otherwise.") But in 2026 the... Read more

0

Slashdot
EditorDavid @ Slashdot · 07/04/2026 16:34 EDT

842,000 American Households Lost Power Today During a Heatwave

As America began celebrating its 250th birthday Saturday, 842,000 homes reported power outages, notes ABC News. Figures from tracking site PowerOutage showed states in America's Northeast and Midwest were impacted by severe weather and extreme heat. That number, which will fluctuate throughout the day as crews work to restore power, is for households, meaning that the number of people impacted by these outages is likely to be much larger... Millions... Read more

0

Most popular sources

  • You see 795 news out of 795.
  • Sources 61 out of 61.
ScienceDaily 0%
Tech Wire Asia 0%
Sifted 0%
ReadWrite 0%
UK Tech News 0%
View sources »

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

10.07.2026 22:18
Last update: 21:50 EDT.
News rating updated: 05:10.

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 © 2026