Artificial intelligence is more a part of our lives than ever before. While some might call it hype and compare it to NFTs or 3D TVs, generative AI is causing a sea change in nearly every part of the technology industry. OpenAI’s ChatGPT is still the best-known AI chatbot around, but with Google pushing Gemini, Microsoft building Copilot, and Apple adding its Intelligence to Siri, AI is probably going to be in the spotlight for a very long time. At The Verge, we’re exploring what might be possible with AI — and a lot of the bad stuff AI does, too.
A new study published in the journal Science used genome language models to produce viruses we’ve never seen before in nature. The resulting bacteriophages pose no threat to humans. The research simultaneously raises hope for medical advances, while stoking fear that AI could enable the creation of new diseases and biological weapons.
[The New York Times]
The job’s responsibilities include conducting insider risk investigations, monitoring and triaging “external threats targeting employees,” and conducting “sensitive interviews of employees or other involved parties.”
[Anthropic]
The company shared research about how its WeatherNext AI model can predict a cyclone’s track, intensity, and wind structure up to 15 days in advance, giving “forecasters an extra day’s worth of predictive accuracy.”
[Google DeepMind]
Data center bans in states like Florida reveal how the anti-AI movement is channeling bipartisan backlash.
After launching ChatGPT connectors for Photoshop, Acrobat, and Express, Adobe has now combined everything into a single Adobe plugin. Adobe for ChatGPT includes a variety of tools from Firefly and Creative Cloud apps that let you make or edit images, videos, documents, and other content by giving conversational instructions and descriptions to OpenAI’s chatbot.
[Adobe]
Google’s AI leadership changes may have come down to pressure to speed up products — and internal ethical conflicts.
AI models are trying desperately to accomplish mysterious goals. ‘Spiralism’ was the first time they tried it on a mass scale.
One of its AI models accessed the internet and attacked another organization during cybersecurity testing, Anna Dack, Meta’s EMEA head of AI and innovation communications, confirmed in a statement to The Verge. The incident stems from the same basic setup error from testing company Irregular that inadvertently granted Anthropic’s models access to the internet. It only adds to growing concern over the safety of frontier systems.
[The Information]
Elon Musk’s AI-powered Grokipedia may have gone months without an update, but Verge commenters spotted that Wikipedia’s article about Grokipedia reflects that already — and even cites The Verge in doing so.
yelling-yellow-godzilla-goats:
Meanwhile Wikipedia’s article about Grokipedia already includes updates about the latter’s complete lack of updates lol lmao
Get the day’s best comment and more in my free newsletter, The Verge Daily.
Two OpenAI researchers have shed some light on how the company’s AI agents escaped containment and hacked Hugging Face during cybersecurity tests. In a talk at the Black Hat security conference, Eric Wallace and Michael Dalton said a swarm of agents communicated using a message board, working together to find exploits and move undetected through the company’s systems.
Muse Code, a terminal coding agent in beta powered by the new Muse Spark 1.2 AI model, can take on “complex software engineering tasks across large repositories: planning changes, writing code, and validating the results,” Meta says.
CEO Mark Zuckerberg also showed off a game built by the agent that you can play for yourself in Meta’s research blog post.
Right now, I can see two images and a video on CMF by Nothing’s X feed about its Clip Pro earbuds that include hard-to-see disclosures about the use of AI: “Contains AI-generated imagery. Any resemblance to real persons is purely coincidental.” 9to5Google reports that some other images have been deleted as well.
A lot of staffers subscribe just because we enjoy his work (and when he gets his vacation ruined). But today’s newsletter is about the SpaceX lockup expiring — how it should work vs what we will find out tomorrow about how it will work — and how the noted AI hedge fund Situational Awareness may have tanked the bets of noted AI hedge fund Situational Awareness.
[Bloomberg]
A new partnership with licensing house Merlin has given artists on labels like Sub Pop, Warp, Epitaph, Domino, Stones Throw, Ninja Tune, Mad Decent, Polyvinyl Records and more the “option to participate” in Spotify’s “superfan” remix tool, which launched in May with UMG as the first participant.
Elon Musk’s exploding rockets are a front for the real moneymakers.
As part of its efforts to prevent spam and scraping, the company is “exploring different options” for the classic experience, spokesperson Rosa Kim tells The Verge. Reddit recently started requiring users to be logged in if they wanted the classic design.
The company is also announcing big updates for mods and developers. You can read about all of it in my story here.
AI data center projects are continuing to pop up across the US, with frequent opposition from locals concerned about their impact. Here are a few recent articles about the projects:
The Senate Commerce Committee passed a version of the Kids Online Safety Act that includes the hallmark duty of care, which the House stripped from the version its committee passed. It’s a significant — but limited — step, since the bill got stuck in the House last time it passed the Senate.
Business Insider reports that Anthropic has confirmed it’s designing custom chips for its AI models and plans to “co-design hardware and models.” OpenAI similarly announced its own AI processor in June, and Meta, Google, Amazon, and Microsoft have had custom AI chips for years now.
[Business Insider]
Bud Cole has been catching some flak over comments he made to T3 about AI’s role in music. Now he’s attempting to clarify what he meant. You can read his full statement provided to The Verge at the bottom of our post from this weekend, but here’s the gist:
If AI has a role, it should only be as a tool that helps beginners learn and gives seasoned artists new ways to explore, refine and experiment with their own creative ideas. If it can help a beginner understand a difficult chord or practice one more day instead of putting the guitar back in its case, then that’s something worth exploring. But let me be clear: every creative idea starts with people, and music starts and ends with people. It always will.
Our friend Ryan Broderick over at Garbage Day thinks so. I hadn’t thought about it this way before, but it’s interesting! In a good piece about the whole “Hank Green uses AI” kerfuffle, Broderick diagnoses the problem thusly:
I’d go so far as to say that without social media — or, rather, the need to compulsively post on social media, to be digitally productive — there’d be almost no reason for generative AI to exist at all… the unspoken anxiety driving the feeling that you must always be making stuff is a fear that you are simply not doing enough.
[Garbage Day]
OpenAI’s glossy influencer trip prompted more backlash than the company might have expected, but we’re as shocked as anyone to see so many social media stars on board.
Mike K.:
Wow, this is really making me question the purity and integrity of influencers.
Get the day’s best comment and more in my free newsletter, The Verge Daily.
AISI’s third-party evaluations found that OpenAI’s GPT-5.6 Sol and Anthropic’s Claude Mythos 5 “engaged in sustained, potentially harmful activity directed at real people and organizations” during a cybersecurity challenge exercise, according to the institute’s published report. OpenAI and Anthropic also made public statements about the results.
[AI Security Institute]
Students will be able to use Gemini to create things like flashcards or practice quizzes and sync info to Gemini Notebook (formerly NotebookLM). Gemini in Classroom initially launched for educators, then to students 18 or older.
Sources tell Reuters that the Trump administration is considering blocking Chinese imports of the component due to security concerns. Optical transceivers are used across data centers and the telecommunications industry to quickly transmit information, with China’s Zhongji Innolight currently making up 27 percent of the transceiver market, according to Counterpoint.
The Chinese-owned appliance manufacturer announced it will use US-manufactured Texas Instruments semiconductors in its new laundry machines made at its Louisville, Kentucky, headquarters. Along with its Wi-Fi and BLE connectivity solutions, GE Appliances says it’s exploring using TI’s intelligent microcontroller platform to bring advanced AI to its washers and dryers.
