Models & Labs · United States

Fired OpenAI researchers say they were let go for 'prioritising safety'

Credited to BBC · bbc.co.uk · 2h ago

Photograph · BBC

Three former OpenAI researchers sacked for allegedly mishandling information claim they were instead let go because they raised safety concerns at the tech firm.

"I believe we were fired for prioritising safety over the near-term interests of OpenAI as a corporation," Mikita Balesni posted on X, external alongside an open letter to the ChatGPT-maker, written with fellow ex-employees.

On 2 October, a spokesperson for the AI firm told the BBC its investigation "confirmed that these individuals mishandled sensitive information outside established company procedures".

OpenAI repeated these claims, external on Friday morning in a note from research leaders.

"We want to be very clear that these decisions were not about raising safety concerns or speaking out," the note said.

"Our internal investigation uncovered a significant breach of trust beyond what's outlined in the letter they published and we stand by the decision to not continue their employment."

The dismissals come amid a growing debate over AI safety and the potential risks advanced AI systems could pose to society.

The joint letter to OpenAI leadership was written by Balesni, Tomek Korbak and Jasmine Wang, who say they were all, until recently, safety or alignment employees at the company.

Alignment is the practice of building human ethical ideas and principles into the technology. In other words, it aims to keep it on track with what humans value.

The former staff members raised concerns that their firings had made employees at the tech company "afraid to speak and operate in ways that, until last week, were an integral part of working at OpenAI".

Reposting the letter, Korbak wrote, external: "For months, I'd been raising safety concerns that we're losing the ability to monitor what AI agents think, one of our best tools for catching when they misbehave.

OpenAI said it was finalising contracts with third-party safety assessors and would announce details in the coming weeks.

"We were not the first to be pushed out of OpenAI under suspicious circumstances," Wang said in a post on X, external.

"Unless the employees take a stand now against this kind of manoeuvre, I am concerned we will not be the last."

Anthropic's Claude Science creates the first complete ultraviolet map of the sky

Credited to The Decoder · the-decoder.com2h ago

Useful

Anthropic's Claude Science just mapped the entire sky in ultraviolet light for the first time. The project is supposed to show how AI can take over tedious data work that researchers would otherwise never get around to. Johns Hopkins astrophysicist Brice Ménard describes the process on the Anthropic website.

UV light reveals dust lit up by starlight, things like clouds around young stars or rings left behind by stellar explosions. A complete UV map didn't exist until now because the ozone layer blocks UV light, so it can only be measured from space. NASA's GALEX mission covered about two-thirds of the sky but skipped bright star-forming regions.

Claude Science coordinates AI agents that downloaded data from multiple space missions, calibrated it, and merged it together. They filled missing areas using inpainting, a technique where a model learns from existing data to reconstruct gaps. In tests, predictions averaged about ten percent deviation from actual measurements. Ad

The map is meant to serve as teaching material. Ménard suspects many scientists have been putting off similar projects that AI could now make possible. Ad

AI News Without the Hype – Curated by Humans

Original · The Decoder

Front Method

Models & Labs · United States

Google brings agentic AI to Gemini, starting with businesses

Credited to TechCrunch · techcrunch.com · 17h ago

At a Google Cloud event on Thursday, the company announced it’s bringing its Gemini AI into the agentic age, with the launch of a unified agent that can not only answer questions but also get things done on the user’s behalf, all from a single interface.

The move comes as AI tools have been moving beyond being just conversational experiences to those that can take ownership of assigned tasks, generate code, schedule meetings, book appointments and travel, and more. It also follows the rise of consumer-facing agents like Meta’s Muse and those that operate over messaging, like Instinct and others, as well as the recent debut of ChatGPT’s Dots.

The company still has a good shot at achieving scale for its agentic efforts — as Google CEO Sundar Pichai pointed out at the event’s start, Gemini today has over 1 billion monthly active users. He also noted that nearly 90% of Fortune 100 businesses now use Gemini Enterprise at work.

Given Gemini’s adoption in the corporate world, Google will initially focus on bringing the agent to businesses before later rolling it out to consumers.

According to Pichai, this will allow the company to solve the “harder problems around security, scale, and performance,” which come with launching powerful agents such as these.

Thomas Kurian, CEO of Google Cloud, explained that the new agent can be given “objectives, not just instructions.” That means it’s able to plan the work, use custom skills and tools, and connect to businesses’ internal systems to accomplish its goals.

By default, the AI will pick the best model to complete the task, but users can also take over to choose a model — including those from third parties, starting with Anthropic’s Claude models. Google said that it will expand the model picker to include open source models and other private models in the future.

The request can include attachments, like files, folders, or other projects designed for specific workstreams, like those that combine files and skills. The agent can connect to the business’ data and systems, like Google Workspace, Microsoft 365, Slack, Jira, Confluence, Git, BigQuery, Databricks, Postgres, Snowflake, and others.

It can also connect and work securely with any Model Context Protocol (MCP) server inside or outside the company’s network.

Users can keep track of what Gemini is doing from a “tasks inbox” interface, where they can see Gemini’s thinking process, delegation of tasks to subagents, loading of special skills, code, and progress.

Notably, the AI will have its own Workspace account, as if it’s just another co-worker. That means it has its own email address and its own context, Google said. It knows who at the company is on which team, their time zones, who needs to approve items, what’s on people’s calendars, and more. Users can call the agent by tagging it, emailing it, sharing with it, or adding it to a group chat. As it takes action, it writes an audit trail attributed to the agent, instead of a person.

The agent will be accessible from iOS and Android mobile devices, Windows and Mac desktops, the command line interface, Google Workspace, Microsoft 365, ServiceNow, and Slack.

The agent’s early testers included sportswear brand On, Shopify, and PayPal, and noted that Gemini Enterprise customers include large businesses like BNP Paribas, Bradesco, Merck, Orange Spain, Santee Cooper, SOMPO, Ulta Beauty, and Wesfarmers.

Google said new flexible spending options, including multi-model orchestration, smart routing, and real-time spend caps, will help businesses keep their enterprise AI costs under better control.