Editor’s Pick

Enlarge / The front page of BreachForums.

The FBI and law enforcement partners worldwide have seized BreachForums, a website that openly trafficked malware and data stolen in hacks.

The site has operated for years as an online trading post where criminals could buy and sell all kinds of compromised data, including passwords, customer records, and other often-times sensitive data. Last week, a site user advertised the sale of Dell customer data that was obtained from a support portal, forcing the computer maker to issue a vague warning to those affected. Also last week, Europol confirmed to Bleeping Computer that some of its data had been exposed in a breach of one of its portals. The data was put up for sale on BreachForums, Bleeping Computer reported.

On Wednesday, the normal BreachForums front page was replaced with one that proclaimed: “This website has been taken down by the FBI and DOJ with assistance from international partners.” It went on to say agents are analyzing the backend data and invited those with information about the site to contact them. A graphic shown prominently at the top showed the forum profile images of the site’s two administrators, Baphomet and ShinyHunters, positioned behind prison bars.

Enlarge / Still images taken from videos generated by Google Veo. (credit: Google / Benj Edwards)

On Tuesday at Google I/O 2024, Google announced Veo, a new AI video synthesis model that can create HD videos from text, image, or video prompts, similar to OpenAI’s Sora. It can generate 1080p videos lasting over a minute and edit videos from written instructions, but it has not yet been released for broad use.

Veo reportedly includes the ability to edit existing videos using text commands, maintain visual consistency across frames, and generate video sequences lasting up to and beyond 60 seconds from a single prompt or a series of prompts that form a narrative. The company says it can generate detailed scenes and apply cinematic effects such as time-lapses, aerial shots, and various visual styles

Since the launch of DALL-E 2 in April 2022, we’ve seen a parade of new image synthesis and video synthesis models that aim to allow anyone who can type a written description to create a detailed image or video. While neither technology has been fully refined, both AI image and video generators have been steadily growing more capable.

Enlarge (credit: BeeBright / Getty Images / iStockphoto)

Infrastructure used to maintain and distribute the Linux operating system kernel was infected for two years, starting in 2009, by sophisticated malware that managed to get a hold of one of the developers’ most closely guarded resources: the /etc/shadow files that stored encrypted password data for more than 550 system users, researchers said Tuesday.

The unknown attackers behind the compromise infected at least four servers inside kernel.org, the Internet domain underpinning the sprawling Linux development and distribution network, the researchers from security firm ESET said. After obtaining the cryptographic hashes for 551 user accounts on the network, the attackers were able to convert half into plaintext passwords, likely through password-cracking techniques and the use of an advanced credential-stealing feature built into the malware. From there, the attackers used the servers to send spam and carry out other nefarious activities. The four servers were likely infected and disinfected at different times, with the last two being remediated at some point in 2011.

Stealing kernel.org’s keys to the kingdom

An infection of kernel.org came to light in 2011, when kernel maintainers revealed that 448 accounts had been compromised after attackers had somehow managed to gain unfettered, or “root,” system access to servers connected to the domain. Maintainers reneged on a promise to provide an autopsy of the hack, a decision that has limited the public’s understanding of the incident.

Enlarge / An image Illya Sutskever tweeted with this OpenAI resignation announcement. From left to right: New OpenAI Chief Scientist Jakub Pachocki, President Greg Brockman, Sutskever, CEO Sam Altman, and CTO Mira Murati. (credit: Ilya Sutskever / X)

On Tuesday evening, OpenAI chief scientist Ilya Sutskever announced that he is leaving the company he co-founded, six months since he participated in the coup that temporarily ousted OpenAI CEO Sam Altman. Jan Leike, a fellow member of Sutskever’s Superalignment team, is reportedly resigning with him.

“After almost a decade, I have made the decision to leave OpenAI,” Sutskever tweeted. “The company’s trajectory has been nothing short of miraculous, and I’m confident that OpenAI will build AGI that is both safe and beneficial under the leadership of @sama, @gdb, @miramurati and now, under the excellent research leadership of @merettm. It was an honor and a privilege to have worked together, and I will miss everyone dearly.”

Sutskever has been with the company since its founding in 2015 and is widely seen as one of the key engineers behind some of OpenAI’s biggest technical breakthroughs. As a former OpenAI board member, he played a key role in the removal of Sam Altman as CEO in the shock firing last November. While it later emerged that Altman’s firing primarily stemmed from a power struggle with former board member Helen Toner, Sutskever sided with Toner and personally delivered the news to Altman that he was being fired on behalf of the board.

Enlarge / A video still of Project Astra demo at the Google I/O conference keynote in Mountain View on May 14, 2024. (credit: Google)

Just one day after OpenAI revealed GPT-4o, which it bills as being able to understand what’s taking place in a video feed and converse about it, Google announced Project Astra, a research prototype that features similar video comprehension capabilities. It was announced by Google DeepMind CEO Demis Hassabis on Tuesday at the Google I/O conference keynote in Mountain View.

Hassabis called Astra “a universal agent helpful in everyday life.” During a demonstration, the research model showcased its capabilities by identifying sound-producing objects, providing creative alliterations, explaining code on a monitor, and locating misplaced items. The AI assistant also exhibited its potential in wearable devices, such as smart glasses, where it could analyze diagrams, suggest improvements, and generate witty responses to visual prompts.

Google says that Astra uses the camera and microphone on a user’s device to provide assistance in everyday life. By continuously processing and encoding video frames and speech input, Astra creates a timeline of events and caches the information for quick recall. The company says that this enables the AI to identify objects, answer questions, and remember things it has seen that are no longer in the camera’s frame.

Enlarge (credit: Getty)

A study analyzing Apple, Microsoft, and SpaceX suggests that return to office (RTO) mandates can lead to a higher rate of employees, especially senior-level ones, leaving the company, often to work at competitors.

The study (PDF), published this month by University of Chicago and University of Michigan researchers and reported by The Washington Post on Sunday, says:

In this paper, we provide causal evidence that RTO mandates at three large tech companies—Microsoft, SpaceX, and Apple—had a negative effect on the tenure and seniority of their respective workforce. In particular, we find the strongest negative effects at the top of the respective distributions, implying a more pronounced exodus of relatively senior personnel.

The study looked at résumé data from People Data Labs and used “260 million résumés matched to company data.” It only examined three companies, but the report’s authors noted that Apple, Microsoft, and SpaceX represent 30 percent of the tech industry’s revenue and over 2 percent of the technology industry’s workforce. The three companies have also been influential in setting RTO standards beyond their own companies. Robert Ployhart, a professor of business administration and management at the University of South Carolina and scholar at the Academy of Management, told the Post that despite the study being limited to three companies, its conclusions are a broader reflection of the effects of RTO policies in the US.

Enlarge (credit: Getty Images | Andriy Onufriyenko)

Billy Restey is a digital artist who runs a studio in Seattle. But after hours, he hunts for rare chunks of bitcoin. He does it for the thrill. “It’s like collecting Magic: The Gathering or Pokémon cards,” says Restey. “It’s that excitement of, like, what if I catch something rare?”

In the same way a dollar is made up of 100 cents, one bitcoin is composed of 100 million satoshis—or sats, for short. But not all sats are made equal. Those produced in the year bitcoin was created are considered vintage, like a fine wine. Other coveted sats were part of transactions made by bitcoin’s inventor. Some correspond with a particular transaction milestone. These and various other properties make some sats more scarce than others—and therefore more valuable. The very rarest can sell for tens of millions of times their face value; in April, a single sat, normally worth $0.0006, sold for $2.1 million.

Enlarge (credit: Getty Images)

On Monday, OpenAI employee William Fedus confirmed on X that a mysterious chat-topping AI chatbot known as “gpt-chatbot” that had been undergoing testing on LMSYS’s Chatbot Arena and frustrating experts was, in fact, OpenAI’s newly announced GPT-4o AI model. He also revealed that GPT-4o had topped the Chatbot Arena leaderboard, achieving the highest documented score ever.

“GPT-4o is our new state-of-the-art frontier model. We’ve been testing a version on the LMSys arena as im-also-a-good-gpt2-chatbot,” Fedus tweeted.

Chatbot Arena is a website where visitors converse with two random AI language models side by side without knowing which model is which, then choose which model gives the best response. It’s a perfect example of vibe-based AI benchmarking, as AI researcher Simon Willison calls it.

Enlarge (credit: Getty Images)

Federal agencies, health care associations, and security researchers are warning that a ransomware group tracked under the name Black Basta is ravaging critical infrastructure sectors in attacks that have targeted more than 500 organizations in the past two years.

One of the latest casualties of the native Russian-speaking group, according to CNN, is Ascension, a St. Louis-based health care system that includes 140 hospitals in 19 states. A network intrusion that struck the nonprofit last week ​​took down many of its automated processes for handling patient care, including its systems for managing electronic health records and ordering tests, procedures, and medications. In the aftermath, Ascension has diverted ambulances from some of its hospitals and relied on manual processes.

“Severe operational disruptions”

In an Advisory published Friday, the FBI and the Cybersecurity and Infrastructure Security Agency said Black Basta has victimized 12 of the country’s 16 critical infrastructure sectors in attacks that it has mounted on 500 organizations spanning the globe. The nonprofit health care association Health-ISAC issued its own advisory on the same day that warned that organizations it represents are especially desirable targets of the group.

Enlarge (credit: Getty Images)

On Monday, OpenAI debuted GPT-4o (o for “omni”), a major new AI model that can reportedly converse using speech in realtime, reading emotional cues and responding to visual input. It operates faster than OpenAI’s previous best model, GPT-4 Turbo, and will be free for ChatGPT users and available as a service through API, rolling out over the next few weeks.

OpenAI revealed the new audio conversation and vision comprehension capabilities in a YouTube livestream titled “OpenAI Spring Update,” presented by OpenAI CTO Mira Murati and employees Mark Chen and Barret Zoph that included live demos of GPT-4o in action.

OpenAI claims that GPT-4o responds to audio inputs in about 320 milliseconds on average, which is similar to human response times in conversation, according to a 2009 study. With GPT-4o, OpenAI says it trained a brand new AI model end-to-end using text, vision, and audio in a way that all inputs and outputs “are processed by the same neural network.”