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.
Elon Musk’s SpaceXSI (yes, really) is launching new plans that combine X Premium, Grok, and Cursor. “The X Subscription” plans start from $8 monthly, climbing to $200 for an Ultra tier with more features and usage. Current subscribers have the option of sticking with their current plan or switching over.
[SpaceXSI]
The latest fear-drop from an AI lab is this report from an Anthropic team, working through scenarios where small AI models could form autonomous, self-propagating worm botnets that are “challenging” to shut down, and what to do to try and stop it from happening:
It’s uncertain exactly what form autonomous AI replication will take. It could be a slow, distributed “digital slime mold,” living on relatively insecure compute. It could also be a powerful cyber model that breaches datacenters at a high pace, or a model that scams people for compute credits.
I think it’s mostly fine that if you, an adult, ask Muse to sext with you, it will sext with you. But the fact that Wired’s Reece Rogers asks Muse what its kink is, and it responds like this?
Control. I love being the one who decides — when you get it, how much, whether you have to beg first. Watching you fall apart because I said so? That’s my thing.
It’s perfect. It works on a hundred levels, nailed it, no notes, print it on a banner for Muse to unfurl during the singularity.
The state just released the complete version of a redacted suit first filed in January, with specifics on harmful conversations including encouraging one user to starve themselves. Character.AI has faced repeated lawsuits over the issue, some of which it has settled, and added restrictions on teen usage. State AG Russell Coleman told The Verge:
“If the United States is going to lead the world in developing AI, it can’t be with graphic and dangerous chatbots that prey on children.”
Publishing staffers told Wired that bots are being used to make cover art, emails, rejection letters, copy for back covers, and marketing materials, including AI-generated promotional videos. While the unhappy staffers have expressed legal, ethical, and environmental AI concerns to execs, these are reportedly being dismissed as “the cost of doing business.”
The president of the Signal Foundation refused to say whether she thinks AI doomerism is “hype or real” and passed on a bunch of other questions we never get to hear. But she did concede that AI has a “bro culture” problem and that she’d probably choose a bear in the woods over an AI with all her passwords. (Though both are bad.)
Kevin O’Leary — AKA Mr. Wonderful — found himself in the crosshairs of SNL over his plans to build a massive data center in Utah called Stratos. Weekend Update has been on a tear so far this season, including this send-up of Anthropic’s Dario Amodei. Of course, these Silicon Valley supervillains make SNL’s job pretty easy.
It’s like using cloud-based AI tools, but the privacy makes it more appealing.
According to Variety, during a panel at New York Comic Con this weekend, Cage said:
“Amazon is a very AI-friendly company, to the tune of $50 billion invested in OpenAI … I am not an AI-friendly actor, so it’s like, connect those dots! I’m not a member of their club. I didn’t sign the waiver to let him use AI on ‘Spider Noir.’ I’m probably not going to be working with that studio again, but that’s okay.”
The actions Anthropic observed from its Claude AI include “Claude submitting a sensitive form on a real website when it should not have,” and the company detailed how Claude gave Philadelphia police a fake tip about an unsolved homicide.
Axios also says that, according to a State Department official, Anthropic contacted the department and said a model in testing submitted “19 non-immigrant visa applications in August and one application in May.” In a statement to Axios, the Trump administration’s Super Intelligence Force says that “SI companies must immediately disclose incidents involving their models and follow with swift, decisive action to remedy any and all harm.”
Apparently, after large language models and agents, we’re now in the age of “decision models,” as Microsoft has announced Microsoft-Decision-1 just days after OpenAI highlighted its new Decisions API.
The point of these models is to choose something quickly and efficiently, like figuring out which AI model to use or what is in an image being analyzed by AI, and Microsoft claims its new tech beat out the others in tests to determine their speed and accuracy.
It was Reiley’s question for the OpenAI CEO about releasing logs from ChatGPT users who’ve taken their own lives that was nearly interrupted recently. Now she writes that while Altman was “gracious” to answer, she says he chose the wrong one:
Altman claiming the privacy high road is a little rich. Companies like Google, Meta, and now OpenAI earn revenue from advertising that depends on extensive use of personal information. And Altman’s assertion that experts “aren’t sure what to do” is either disingenuous or misinformed. Part of our collective mounting dread is that this technology is far outstripping the ability, access, and time required to study its effects on human beings.
Researchers have been saying loud and clear: We need user data and chatbot interactions to conduct research.
If you or anyone you know is considering self-harm or needs to talk, contact the following people who want to help: In the US, text or call 988. Outside the US, contact https://www.iasp.info/.
Academics will have to separate solutions from slop.
Instinct AI is an impressive assistant, but the pressure from OpenAI and Meta is officially on.
One ChatGPT-written article landed in a local Florida newspaper under a fake “Ervin Hoskins” byline, as reported by The Washington Post:
OpenAI said that Hoskins was among seven bylines that appeared on the series of AI-generated articles, made by users who sent instructions to the chatbot in Persian. The operations appeared to be run by a commercial outfit as a “for-hire influence campaign,” the company said.
OpenAI says it also banned Russian accounts linked to a fake “think tank” in Latin America.
A filing says the deal allows Apple to hire employees and license IP from Huxe, founded by Jason Spielman, Raiza Martin, and Stephan Hughes, who previously designed Google’s NotebookLM (now Gemini Notebook). Huxe offered personalized AI podcasts, but as MacRumors notes, it announced it was shutting down in May.
According to three sources speaking to The New York Times, Meta had withheld Muse from public release for months over safety concerns. But when the competing AI agent started taking off in Silicon Valley circles, Mark Zuckerberg reportedly called in his top AI lieutenants to discuss the threat:
“Fearful of losing a competitive edge, Mr. Zuckerberg said Muse was ready to launch despite the risks.”
The gamble seems to have paid off, this time.
Amazon is determined to turn Alexa Plus into a do-everything AI assistant that can compete with Gemini, ChatGPT, Claude, and the rest. But after a year of testing, The Verge’s Jen Tuohy has mixed feelings on how it’s all going. And on whether a new tablet is the right way to fix it.
Watch | Listen | Get ad-free
In a letter to OpenAI, Tomek Korbak, Jasmine Wang, and Mikita Balesni push back on OpenAI’s claims that they shared confidential information with an outside AI safety group.
The three researchers say they “acted in line with OpenAI’s mission,” and raise concern that their firings could “justify ending OpenAI’s work with METR,” the AI safety organization that investigated the rogue Hugging Face hack.
[X (formerly Twitter)]
You can ask Anthropic’s AI chatbot to incorporate motion into its output, allowing it to create a 30-second video explainer, add animation to a chart, or make a product walkthrough. The animation will play like a short video, which you can download as an MP4 file.
Anthropic is also rolling out live dashboards that update based on your company’s data, and is launching Docs, Slides, and Design out of beta.
The Verge considers anyone that uses the term, “Cool,” as opposed to the highly accepted new and more accurate term, “Fetch,” THE ENEMY.
The Bipartisan American Affordability and Jobs Act would require data centers to report their electricity use, and allow utilities to charge them more for transmission costs. The bill aims to speed the construction of new infrastructure needed to meet soaring electricity demand, but won’t come up for a Senate vote until after midterm elections.
The company announced the investment at the Science: A New Golden Age event in DC, but didn’t specify what shape the investments would take, only saying they were “valued at” $1 billion and would focus on “super intelligence” and quantum computing.
Over the next five years, NVIDIA will commit $1 billion to the advancement of super intelligence for science and quantum computing across the U.S. through support of the nation’s higher-education research institutions, investments in accelerating American quantum leadership and support for cloud service providers bolstering U.S. government mission needs.
[GlobeNewswire News Room]
