FBI Seizes Domains and DoorDash Launches AI

Are roll batteries the same? That's like asking ev soccer players are the same Take Messi, the most decorated player ever Is there any other player who has achieved that? No, just him. Now take Duracl. Is there any other battery with power boost ingredients inside? No, just Duracl. Remember, goats only trust goats because they're built different. and Messi only trusts Duracl Welcome to the Tech R Rite Home for Thursday, june eleventh, twenty six. I'm Brian Maella. today, anthropic backtracked on secretly degrading Fable five for AI researchers. Open AI considers drastic token price cuts to start a price war. Dario Amadai calls for FAA style AI regulation. The FBI seized fake Chinese consulting domains, and Doorash is launching AI ordering Here's what chimmiss today in the world of tech Think you have to build your own search engine scraping infrastructure. Thank again, folks. SRP API can take care of all your search engine scraping needs. It lets AI products access real time web search data programmatically No scraping infrastructure required. Their APIs offer structured JSON data from all the top search engines, including Google, Amazon, and YouTube Tech companies already use SERP API to power their AI agents, their market intelligence tools, and their automated research workflows See how it can help you and your team SRP API provides two hundred and fifty free searches per month Go check it out at SRPapPi dot com That's SRP SRpapPi dot comot Anthrapic has been facing a bunch of backlash over the last few days. People are pissed that Fable five is going to be switching to full metered use in a matter of days That sound you hear is a million vibe coders. rushing to get their projects done before it becomes too expensive to use, Fable Cybersecurity researchers are complaining that Claude Fabable's guardrails are too strict rejecting innocuous tasks like recording and reading blog posts or performing code reviews. Microsoft is restricting employee usage of Cloud Fable five because of Anthropic's new thirty day retention requirements in terms of the prompts even. And some people are arguing that anthropic secretly limits Claude's usefulness for LLM development thereby strengthening the argument that Anthropic is using AI safety to justify monopolistic behavior Anthropic is responding to some of these concerns by today backtracking on its decision to quietly limit Fable five's ability to develop LLMs, saying requests will be visibly falling back to Ous four point eight with the underline there on visibly. quoting Wired Athropic released Claud of Fable five a version of its latest AI model with additional safety guardrails designed to prevent misuse earlier this week. S of the safeguards Anthropic decided were surprising The company said, it would reroute users who ask questions about cybersecurity, biology, or chemistry to a less capable AI model to reduce the chances of someone using the advanced AI to carry out a cyber attack or build a bioeapon. But for researchers trying to use Claud Fable five for frontier AI development, Anthropic outlined a different approach. The firm would deliberately degrade the model's performance in ways that were invisible to the user The move would effectively sabotage researchers trying to use ClOD to train competing AI models, which Athropic explicitly bans in its terms of service Anthropic now says it's changing course and that Claud Fable five's safeguards for AI development will be visible to users. If the company suspects a user is trying to use Claud to build a highly capable AI, it will alert them that it's either refusing the request or rerouting the user to a less capable model. Anthropic reversed the policy after it received fierce backlash from the AI research community. Anthropic has already taken steps to limit competitors from using Cloud to build closed and open source AI models, but critics say that quietly Degrading the model's performance for certain users went a step too far. Claud's codating agent has become a favord tool among developers, including those working on open source AI research projects, and researchers tell Wired that the company's latest policy would have led to a troubling future in which only a handful of leading AI labs could perform advanced AI research Dean Ball, a senior fellow at the Foundation for American Innovation and a former advisor to the White House on AI, wrote in a post on X that degrading performance on ML research without telling the user is shockingly hostile and a terrible look He continued in another post that the secret sabotage policy undermines Anthropic's overall stance because it limits AI researchers from collaborating on AI safety. It felt like Anthropic was saying to the public, We don't trust anybody else to do AI research. We are the only ones who have to do AI research says Brown, research lead at the open source AI startup Prime Intellect It feels a bit like they are starting to pull the ladder up behind them, end quote Brown said the policy would also have left developers in the dark about whether they were violating anthrop' rules since the company wouldn't alert them when its safeguards were triggered He added that the restrictions could have had widespread consequences. For example, he pointed to the growing ecosystem of third party evaluation firms that test frontier models for safety, performance and reliability, work that could have been hindered if Anthropics secretly degraded its model Anthropics said it implemented the measures because Claud has become increasingly effective at accelerating AI research. In a recent blog post, the company said it is concerned that AI could improve its capabilities faster than society can adapt to them. Anthropic argued that it would be good for the whole world to have the option to slow or temporarily pause frontier AI development to enable societal structures and alignment research to keep up These safeguards prevent foreign adversaries from using our most capable models in ways that pose severe safety risks The U.S. and its allies hold an edge in frontier chips and the highly optimized software that runs them at full potential, the company said in a statement to Wired. These safeguards ensure Claud isn't used to erode that advantage by optimizing chips developed by those adversaries, for example, in deciding whether to make them visible or invisible, we faced a choice A hidden safeguard is harder to probe and work around. This means the safeguards can be targeted much more narrowly and quote Thethropics says that because the safeguard around AI development is now visible, it needs to cast a wider net, meaning more benign requests may trigger its safeguards. The company says it's working to make its classifiers more precise as quickly as possible, end quote. And quoting the Verge. In Fabable S systemstem Card, a public document AI developers released to explain how a system works, Athropic said it would handle queries it believed were distillation attempts by altering and degrading the model's answers directly. Users would not be notified that they had triggered the safety measure or informed that the response had been changed.Qote, Visible safeguards can be probed, so they have to be robust, which takes time to get right, Anthropic wrote Invisible safeguards can be targeted more narrowly, allowing us to ship quickly with very few false positives. We went with invisible safeguards for this reason, and that was the wrong trade off You should have visibility into the safeguards we have in place and why We're sorry for not getting the balance right, and quote Sources tell the journal that OpIe is considering drastically lowering its prices, the prices it charges users for tokens in anticipation of similar cuts the startup expects anthropic to soon make Given this and the recent news that Sam Maltman is actively asking for the U.S. government to come in and take a big stake in open AI, This is making people whisper about whether or not Open Aye is getting desperate or is a price warar just being set up because everybody's about to go public Because you know, nobody can afford the current usage of their users. So you wouldn't think anybody could afford any sort of drastic price cuts at any point coting the journal Business executives have begun to bulk at the high prices for AI usage. Open AI Chief executive Sam Altman said at a recent event that costs have become, quote, a huge issue. I think we'll have a lot of ways we can help people get more value for less spend, he said Drastic price cuts could potentially erode the profit margins of both companies, which already lose billions of dollars because of the enormous cost for computing resources needed for AI systems to process queries and carry out tasks Open AI is trying to catch up with its younger rival in the race to win enterprise customers that are paying large amounts of money for AI tools that can improve workplace productivity Andthropics revenue recently surged after its coding tool clawed code went viral among software engineers and the five year old startup surpassed open eyes valuation for the first time OpenIe has since made its own coding tool, Codex a focus of the company Some corporations poured so much money into Anthropics's products that their leaders are now seeking to rein in spending eararlier this year, an Uber executive said the company had maxed out its twenty twenty six budget for aggentic or autonomous AI use And another company leader said last month that it was difficult to link AI coding productivity improvements to new customer features. Such comments from many executives have triggered a debate within Silicon Valley about token maxing or the practice of using as many tokens as possible to boost productivity including in ways that don't generate returns on investment A price war it would be an early test of the strength of both companies' business models ahead of hotly anticipated public listings. OpenIe and Antropic have captured the majority of revenue from new AI products powering their rise, but an underlying risk that investors have long identified is the interchangeability of their products and the ease with which customers can abandon one or the other Opening Iye confidentially filed for an IPO earlier this week, following in the footsteps of Antthropic In a recent sllack message to employees, Sam Altman said the company plans to go public quote within the next year, end quote AI is uncharted territory, and many leaders are trying to navigate through without a guide to help them. That's why Morning Bw created the Intelligence Shift, a new podcast with PWC. It's all about how AI is fundamentally changing different industries host Dan Priest sits down with people who work with AI on a daily basis. Together, they discuss real stories, real strategies, and real takeaways for leaders. Get guidance from industry experts. listen to the intelligence shift. Whver you get your podcast. Spring has sprung, folks. As the weather warms up, it's a great time to take your routine outside. To keep cool as the days warm up, check out Viori clothing. Viori is designed to keep you looking great both inside the gym and out in the world Biori's products are built for comfort and performance. They're designed to move with you through every part of your day, from your morning workout to your evening commute and kicking back at home. Get twenty percent off your first purchase at Viori dot com slash tech brew and discover the versatility of Viori clothing.s Viori dot com slash tech brerew. Exclusions apppply, visit the website for full terms and conditions So good, so good New summer arrivals are at Nordstrom Rack stores now. Get ready to save big with up to sixty percent off brands like Rag and Bone, Levi's, Adidas, and Free People. Join the Nordy Club to unlock exclusive discounts, shop new arrivals first, and more. Plus, buy online and pick up at your favorite rack store for free Great brands, great prices. That's why you rack Meanwhile, Dario Amodai keeps playing the please regulate us card oting Venture Beat. In a sweeping new essay titled Policy on the AI Exonential, anthropic co founder and CEO Dario Amadeai publicly calls for new government regulations governing the release of powerful AI models specifically comparing AI as an industry to commercial aviation, which follows regulations enforced by the U. S. Federal Aviation Administration or FAA, arguing This is necessary to maintain public safety as AI capabilities and potential misuses grow. Alongside the essay, Anthropic released two comprehensive policy roadmaps, an advanced AI framework targeting catastrophic model risks and an economic policy framework addressing AI driven labor displacement backed by three hundred fifty million dollars in new funding Amadai explicitly compares the necessary AI regulation regime to the Federal aviation administration, stating Frontier AI models like airplanes should be required to go through technical testing and auditing and their release should be blocked or reversed as a threat to public safety if they do not meet high standards of safety. The company is proposing that models trained using more than T to the twenty fifth floating point operations or flops or developed by companies with over five hundred million dollars in AI revenue or one billion dollars in AI R andD, must undergo mandatory third party testing Perhaps the most sobering aspect of the announcement is Anthropics's economic policy framework The company is publicly acknowledging that if AI achieves its predicted capabilities, it will act as a general substitute for labor rather than just a productivity tool Emodai frames this bluntly. The key challenge in such a world won't be incentivizing growth, but finding a way for everyone to share in the benefits, end quote This up As said, Anthropic is committing three hundred fifty million dollars to address economic disruption, two hundred million dollars for an economic Futures research fund to pilot public policy solutions, and one hundred fifty million dollars for a national fellowship program The framework actively plans for scenarios where AI drives unemployment from five to ten percent or even unprecedented levels, advocating for policies like wage insurance, universal basic income and sovereign wealth models, end quote The FBI has seized thirteen domains allegedly tied to fake consulting firms that sought information from U.S. government and military employees for suspected Chinese agents Qoting Reuters. These fake firms target people via job listings for consulting or analyst roles, andid pressure applicants for exclusive or insider information, the department said in a statement The announcement of the domain seizures came a week after the United States, Britain, and other fiveive ey'es intelligence alliance countries warned of China aggressively and increasingly using job platforms to target people for information A spokesperson for the Chinese emmbassy in Washington said that the allegation of so called Chinese espionage threat is entirely fabricated and constitutes malicious slander. We strongly condemn this Allegations that Chinese intelligence is using phony consultancies to recruit American and other Western officials are not new Reuters reported in march twenty twenty five that a similar network of fake consulting firms was attempting to enlist federal employees who had been recently fired as part of President Donald Trump's downsizing and reshaping of the government The FBI and the National Counterintntelligence and Security Center published a short film in september twenty twenty, dramatizing the case of former CIA officer Kevin Mallory. who in twenty nineteen was sentenced to twenty years in prison after being convicted on charges of conspiring to transmit U. S. defense secrets to China Mallory had initially been recruited for foreign policy consulting via social media, according to court records, and his story is prominently featured on an FBI website warning of Qote virtual espionage using similar tactics, end quote YouTube is rolling out a new in app messaging system for sharing videos and having one on one conversations ording nine to five Goog Once available, the top right corner of the YouTube app will show a new messaging icon This system lets you share videos, shorts or live streams and have one on one conversations. One nuance of this capability is how you can only talk with a person after inviting them using a URL. That invite link valid for seven days must be sent using a third party chat platform. It's designed to let you have conversations with friends and family that you already know rather than the broader YouTube audience. In testing last year, YouTube said this was a top feature request and something it previously offered until removing in twenty nineteen After the recipient clicks that link, you send them, they have to allow messaging or select not now. This makes YouTube's messaging feature a bit more narrow and just focused on video sharing. Additionally, unlisted video links can be shared, but not private ones Messaging and YouTube is only available for users eighteen and older with community guidelines applied to all shared content and messages, messages can be unsent, while there's the ability to block users from messaging you again and reporting conversations, end quote And fininally today, why wait for the OSs to get agentic when you can just get agentic yourself Putting CNBC. Doorash is bringing artificial intelligence deeper into the user experience, allowing customers to order food and make reservations with photos and prompts The company on Thursday announced a new chatbot called Ask Doorash, which is launching in select markets for grocery shopping and food delivery. Doorash plans to add reservations and additional US cities in the coming weeks Gig economy companies are in a race to add AI into their apps as the rapid development of agentic tools changes how consumers use the interternet and mobile devices. DoorDash, along with Uber and Instacart, are rolling out new services to keep from getting left behind in a sector that's become a testing ground for AI agents Doorash launched AI powered tools for merchants in May and is betting on autonomous tech like delivery robots. Earlier this year, Uber launched its own AI cart assistant that uses photos and prompts to build grocery lists. and late last year, Instacart introduced AI tools for grocers The stakes are high for DoorDash, which is in the middle of a massive investment cycle that involves the creation of a unified tech platform to house all its brands following a string of big acquisitions purchases include a one point two billion dollars deal for restaurant booking platform seven rooms and the nearly four billion dollars acquisition of deelivereoo and quote. Nothing more for you today. talkalk to you tomorrow. Job site Moved. your crew needs four more rooms. Tomorrow. If you booked on your own, you could be left high and dry. But you booked with Engine, found the right hotel with one click, and avoided cancellation fees. because Engine lets you do that, too. Yes change, no charge. Engine Free to join, Flexible when plans change. Teams traravel spend managed. headad to engine dot com slash flex to get started