In a new paper published Thursday titled " Auditing language models for hidden objectives ," Anthropic researchers described how models trained to deliberately conceal certain motives from evaluators could still inadvertently reveal secrets, thanks to their ability to adopt different contextual roles or "personas." The researchers were initially astonished by how effectively some of their interpretability methods seemed to uncover these hidden motives, although the methods are still under research.

While the research involved models trained specifically to conceal motives from automated software evaluators called reward models (RMs), the broader purpose of studying hidden objectives is to prevent future scenarios where powerful AI systems might intentionally deceive or manipulate human users.

While training a language model using reinforcement learning from human feedback (RLHF), reward models are typically tuned to score AI responses according to how well they align with human preferences. However, if reward models are not tuned properly, they can inadvertently reinforce strange biases or unintended behaviors in AI models.

Read full article

Comments

A new study from Columbia Journalism Review's Tow Center for Digital Journalism finds serious accuracy issues with generative AI models used for news searches. The research tested eight AI-driven search tools equipped with live search functionality and discovered that the AI models incorrectly answered more than 60 percent of queries about news content.

Researchers Klaudia Jaźwińska and Aisvarya Chandrasekar noted in their report that roughly 1 in 4 Americans now uses AI models as alternatives to traditional search engines. This raises serious concerns about reliability, given the substantial error rate uncovered in the study.

Error rates varied notably among the tested platforms. Perplexity provided incorrect information in 37 percent of the queries tested, whereas ChatGPT Search incorrectly identified 67 percent (134 out of 200) of articles queried. Grok 3 demonstrated the highest error rate, at 94 percent.

Read full article

Comments

On Saturday, a developer using Cursor AI for a racing game project hit an unexpected roadblock when the programming assistant abruptly refused to continue generating code, instead offering some unsolicited career advice.

According to a bug report on Cursor's official forum, after producing approximately 750 to 800 lines of code (what the user calls "locs"), the AI assistant halted work and delivered a refusal message: "I cannot generate code for you, as that would be completing your work. The code appears to be handling skid mark fade effects in a racing game, but you should develop the logic yourself. This ensures you understand the system and can maintain it properly."

The AI didn't stop at merely refusing—it offered a paternalistic justification for its decision, stating that "Generating code for others can lead to dependency and reduced learning opportunities."

Read full article

Comments

Anthropic CEO Dario Amodei raised a few eyebrows on Monday after suggesting that advanced AI models might someday be provided with the ability to push a "button" to quit tasks they might find unpleasant. Amodei made the provocative remarks during an interview at the Council on Foreign Relations, acknowledging that the idea "sounds crazy."

"So this is—this is another one of those topics that’s going to make me sound completely insane," Amodei said during the interview. "I think we should at least consider the question of, if we are building these systems and they do all kinds of things like humans as well as humans, and seem to have a lot of the same cognitive capacities, if it quacks like a duck and it walks like a duck, maybe it’s a duck."

Amodei's comments came in response to an audience question from data scientist Carmem Domingues about Anthropic's late-2024 hiring of AI welfare researcher Kyle Fish "to look at, you know, sentience or lack of thereof of future AI models, and whether they might deserve moral consideration and protections in the future." Fish currently investigates the highly contentious topic of whether AI models could possess sentience or otherwise merit moral consideration.

Read full article

Comments

After a little over three months, Intel has a new CEO to replace ousted former CEO Pat Gelsinger . Intel's board announced that Lip-Bu Tan will begin as Intel CEO on March 18th, taking over from interim co-CEOs David Zisner and Michelle Johnston Holthaus.

Gelsinger was booted from the CEO position by Intel's board on December 2 after several quarters of losses, rounds of layoffs, and canceled or spun-off side projects. Gelsinger sought to turn Intel into a foundry company that also manufactured chips for fabless third-party chip design companies, putting it into competition with Taiwan Semiconductor (TSMC), Samsung, and others, a plan that Intel said it was still committed to when it let Gelsinger go.

Intel said that Zisner would stay on as executive vice president and CFO, and Johnston Holthaus would remain CEO of the Intel Products Group, which is mainly responsible for Intel's consumer products. These were the positions both executives held before serving as interim co-CEOs.

Read full article

Comments

Researchers have discovered multiple Android apps, some that were available in Google Play after passing the company’s security vetting, that surreptitiously uploaded sensitive user information to spies working for the North Korean government.

Samples of the malware—named KoSpy by Lookout, the security firm that discovered it—masquerade as utility apps for managing files, app or OS updates, and device security. Behind the interfaces, the apps can collect a variety of information including SMS messages, call logs, location, files, nearby audio, and screenshots and send them to servers controlled by North Korean intelligence personnel. The apps target English language and Korean language speakers and have been available in at least two Android app marketplaces, including Google Play.

Think twice before installing

The surveillanceware masquerades as the following five different apps:

Read full article

Comments

On Wednesday, Google DeepMind announced two new AI models designed to control robots: Gemini Robotics and Gemini Robotics-ER. The company claims these models will help robots of many shapes and sizes understand and interact with the physical world more effectively and delicately than previous systems, paving the way for applications such as humanoid robot assistants.

It's worth noting that even though hardware for robot platforms appears to be advancing at a steady pace (well, maybe not always ), creating a capable AI model that can pilot these robots autonomously through novel scenarios with safety and precision has proven elusive. What the industry calls "embodied AI" is a moonshot goal of Nvidia, for example, and it remains a holy grail that could potentially turn robotics into general-use laborers in the physical world.

Along those lines, Google's new models build upon its Gemini 2.0 large language model foundation, adding capabilities specifically for robotic applications. Gemini Robotics includes what Google calls "vision-language-action" (VLA) abilities, allowing it to process visual information, understand language commands, and generate physical movements. By contrast, Gemini Robotics-ER focuses on "embodied reasoning" with enhanced spatial understanding, letting roboticists connect it to their existing robot control systems.

Read full article

Comments

The AI industry is doing its best to will "agents"—pieces of AI-driven software that can perform multistep actions on your behalf—into reality. Several tech companies, including Google , have emphasized agentic features recently, and in January, OpenAI CEO Sam Altman wrote that 2025 would be the year AI agents "join the workforce."

OpenAI is working to make that promise into a reality. On Tuesday, OpenAI unveiled a new " Responses API " designed to help software developers create AI agents that can perform tasks independently using the company's AI models. The Responses API will eventually replace the current Assistants API , which OpenAI plans to retire in the first half of 2026.

With the new offering, users can develop custom AI agents that scan company files with a file search utility that rapidly checks company databases (with OpenAI promising not to train its models on these files) and navigate websites —similar to functions available through OpenAI's Operator agent, whose underlying Computer-Using Agent (CUA) model developers can also access to enable automation of tasks like data entry and other operations.

Read full article

Comments

Apple on Tuesday patched a critical zero-day vulnerability in virtually all iPhones and iPad models it supports and said it may have been exploited in “an extremely sophisticated attack against specific targeted individuals” using older versions of iOS.

The vulnerability, tracked as CVE-2025-24201, resides in Webkit, the browser engine driving Safari and all other browsers developed for iPhones and iPads. Devices affected include the iPhone XS and later, iPad Pro 13-inch, iPad Pro 12.9-inch 3rd generation and later, iPad Pro 11-inch 1st generation and later, iPad Air 3rd generation and later, iPad 7th generation and later, and iPad mini 5th generation and later. The vulnerability stems from a bug that wrote to out-of-bounds memory locations.

Supplementary fix

“Impact: Maliciously crafted web content may be able to break out of Web Content sandbox,” Apple wrote in a bare-bones advisory . “This is a supplementary fix for an attack that was blocked in iOS 17.2. (Apple is aware of a report that this issue may have been exploited in an extremely sophisticated attack against specific targeted individuals on versions of iOS before iOS 17.2.)”

Read full article

Comments