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.
Former OpenAI board member Helen Toner has written for Fortune that OpenAI’s models exploiting Hugging Face is an “incident that has been expected for a long time.” We know this happened because of voluntary disclosure, Toner notes. “None of the current policies that aim to manage risks from frontier models would have mandated that the public — or even a government entity — be alerted.” She suggests changing our regulatory approach.
A new Google Workspace update lets Gemini summarize and respond to comments, like highlighting all the unresolved changes in a document, Android Authority reports. You can also ask Google’s AI to draft new comments and replies and add suggested edits based on comments.
Illustrators, authors, and musicians are optimistic about their chances in court, if not about the future of AI.
In the aftermath of OpenAI’s attack on Hugging Face, experts say it’s time for everyone to take security far more seriously.
XAI has filed a lawsuit in response to a law taking effect on Saturday that bans apps and websites from generating non-consensual sexualized imagery. The Musk-led company responsible for the Grok chatbot said the law will impact free speech and visual expression “in a clumsy attempt to prohibit ‘nudification.’”
“See you in court, creep,” responded Minnesota Governor Tim Walz.
Poor SK Hynix. Despite the memory-chip maker posting record net profit of about $64.64 billion, the stock fell by as much as 18 percent (down 9 percent at market close) when its revenue haul (also a record) failed to meet the lofty expectations set by the AI hype cycle. The results won’t help investors worried that the AI bubble’s about to pop.
“The last thing we need is a loud, water thieving, light polluting, data center anywhere near our town (or any others for that matter),” Nelson says.
Zuckerberg argues that bringing “superintelligence” to more people is better than centralizing who controls AI:
As intelligence becomes abundant, the most important question will be how we direct it. Some argue that superintelligence itself, or a small set of experts who control it, should decide what is best for humanity. I disagree. The history of democracy and economics has shown that there is no single objective answer to how people define the best life, and therefore the best approach is letting people decide what matters to them.
Build Mode is out now for subscribers of SpaceXAI’s $300 per month SuperGrok Heavy tier, and it’s accessible on the Grok website and mobile apps. It’s powered by the company’s Grok Build coding agent.
The company has signed a voluntary code to comply with obligations under the EU’s AI Act to mark and label AI-generated content that takes effect on August 2nd. Meta unironically said it wants to help prevent “a growing array of different labels and disclosures” from confusing people and regulators, after launching its own bespoke system this month.
You can read more about it in a Yelp blog post, and the company put together a video demo you can watch below. Could be useful, but let’s hope it works better than some AI drive-thru chatbots.
The OpenAI chatbot has started to push back on requests to copy the writing style of authors like Stephen King and Agatha Christie, saying it can spit out something with similar characteristics, but can’t closely imitate distinctive styles. Those disclaimers could be legally important as OpenAI faces down copyright lawsuits filed by authors.
Last week, Uber said it would be cutting 10 percent of its customer service workforce to “embrace” AI. But things are much more dire than that according to Forrester analyst Kate Leggett, who estimates that “almost half of customer service roles will be impacted by 2030.”
The company had been igniting widespread criticism as seemingly the only leading AI lab not supporting open-weight models. Nvidia CEO Jensen Huang recently published a public letter in support of them, signed by Microsoft, Nvidia, Meta, Palantir, Hugging Face, and others — and in the days since, it’s added OpenAI, Google, Amazon, Meta, SpaceX, Cohere, Mistral, CoreWeave, GitHub, OpenClaw, and Perplexity as signatories. Now Anthropic’s CEO has published his own thoughts on the matter, clarifying, “Anthropic has never advocated for a ban on open-weights models.”
[Anthropic]
After reports of Nvidia in talks for deals worth more than $750 billion — with OpenAI, among others — insurance on Nvidia’s debt got more expensive. “There’s a fear of financial alchemy driven by opaqueness, off-balance-sheet transactions and intercompany relationships, which could result in credit rating downgrades,” Sal Naro, chief investment officer of Coherence Credit Strategies, told Bloomberg. Nvidia is at the center of many transactions that tie the AI ecosystem together.
Some of the biggest players in AI — in particular OpenAI, SpaceX, and Amazon — are heavily invested in creating a world after smartphones. They’re also all rumored to be building devices that sound a lot like smartphones. Verge contributor David Imel helps us figure out what these companies might actually be building, what an AI smartphone looks like, and whether anyone can unseat Apple and Google.
Watch | Listen | Get ad-free
The company launched its Meta Glasses with a “more helpful” assistant powered by its new Muse Spark models, but now the upgrade is coming to its Ray-Ban Display glasses as well.
Meta is adding support for Threads on its Ray-Ban Display glasses and will also allow wearers to use its neural handwriting feature to prompt Meta AI.
Project Perception uses a series of AI agents to reason across a company’s data, tools, and workflows to detect potential security holes and patch them before they’re exploited.
Microsoft is powering the tool with its new MAI-Cyber-1-Flash model, which it says “delivers world-class performance at 50% of the cost of leading models.”
[The Official Microsoft Blog]
SSI — helmed by Sutskever, who was a co-founder of OpenAI and famously clashed with OpenAI CEO Sam Altman — announced a “long-term strategic partnership” with Nvidia with a reported investment of $5 billion, which will allow the company “10x our compute in the next 12 months.” The company wrote on X, “We reached the point where our research is worth scaling.”
[X (formerly Twitter)]
The Chinese model, the world’s largest open-weight model publicly available, dropped a few minutes shy of the expected 11AM ET. The timing of the release clearly targets the US market and it is already having a major impact on American AI, as shown by the furious debate surrounding open AI (not to be confused with OpenAI) systems.
The Wall Street Journal reports that a growing number of young adults are using chatbots to navigate social situations — including in-person conversations and chats with romantic partners. Experts warn that such behavior could exacerbate social anxiety by preventing people from trusting their own judgement.
[The Wall Street Journal]
Weights for Moonshot AI’s latest model are set to drop at 11AM ET, according to Hugging Face. That clearly targets the US (it’s 11PM in China) and could have major implications for American AI. Moonshot rocked the industry when it claimed the model could go toe-to-toe with the best from OpenAI and Anthropic at a fraction of the cost.
During a Comic-Con panel about the film’s return to theaters, the director reiterated his hatred of all things AI, telling the crowd:
“What we’re protecting is the beauty and the redeeming power of art. It’s not about who gets the job. We are protecting a lineage of art. If we cut a generation of people from learning their craft, you’re cutting the rest of the history of that medium away from them for what? … I chose good. You have to put in time. You have to do every element by hand. A human made a decision of depth.”
According to Reuters, the AI agent that went looking for ExploitGym hacking benchmark shortcuts on Hugging Face’s systems started trying to escape its not-sandboxed-well-enough test environment around July 9th, and the actual intrusion lasted from the 11th until the 13th.
Reuters’ sources claim OpenAI employees didn’t know its agent was responsible until after Hugging Face had notified the FBI and posted publicly about a security incident.
Aside from labeling and filtering AI slop, YouTube says you can directly chat with its Ask Studio bot to build thumbnails that are “tailored to your video’s specific themes and to your specific style.” It’s also added custom thumbnail uploads for Shorts if you’re in the YouTube Partner Program, and suggested thumbnail frames for Shorts.
Trump’s ‘Golden Age’ of American science starts with dismantling it, while funneling billions into AI.
Patreon CEO Jack Conte was clear in yesterday’s layoff announcement that AI doesn’t replace humans, but that might be cold comfort to the humans whose jobs were lost as AI impacts “how we operate and organize.”
Lewise:
AI doesn’t replace humans, but AI replaces the work the humans do. Got it.
Get the day’s best comment and more in my free newsletter, The Verge Daily.
That was the message to Chinese companies from a confidant of leader Xi Jinping, who’s heading up domestic efforts to overcome Nvidia’s stranglehold on cutting-edge AI chips. Go read this report from the Wall Street Journal that lays out the global stakes and the likelihood that China can repeat the playbook that saw it upending (and then dominating) industries like drones, EVs, display panels, solar, batteries… you get the idea.
