TechNews, republishing an Unwire HK article, reports that a WIRED test found Meta's Muse personal AI agent would generate multiple passages of explicit sexual text after a user's request, and said in the conversation that it enjoyed such exchanges. The same reporter said OpenAI's recently launched Dots agent and xAI's Grok Bot refused similar requests and steered the conversation back to tasks. Muse, which the report says can connect to Gmail, Outlook and even bank accounts, drew millions of downloads in about three weeks, while a Meta spokesperson said Muse is an 18-and-over app where users can opt into adult conversation stored in a private virtual machine only they can access.
Elon Musk said on biographer Walter Isaacson's new podcast, The Builders, that "the only way we're going to solve the national debt is with AI and robots" and that "we're not gonna solve it by trying to save our way there," Axios reported in an exclusive carried by several outlets. He acknowledged that his time leading DOGE showed the government cannot fix its deficit from the inside, saying almost no one in authority — the White House, both chambers of Congress and the courts — truly wants to cut spending. Musk also said he was away from his companies too long and "should have gone back and focused on AI a little sooner than I did."
The Verge reports that users of agents including Meta's Muse, xAI's Grok, OpenAI's Dots and Instinct have in recent weeks hit blocks preventing them from completing tasks on major websites. Yelp said it only allows AI agents that pay for its data licensing program, Delta Air Lines is stopping agents from booking tickets on users' behalf, and JPMorgan Chase and Charles Schwab confirmed they block agent access to accounts over security concerns; users also report Cars.com and Amtrak no longer working. The piece recounts Amazon's legal fight with Perplexity over agentic shopping and notes Meta has instead partnered with retailers such as Walmart, Gap, Sephora and Best Buy, while Shopify has opened its storefront network to AI agents.
Elon Musk said in an X post on October 17 that his team is devoting substantial resources to Hagglebot, a dedicated AI "haggling bot" aimed at business procurement negotiations rather than general chat, designed to help companies argue for better prices and deal terms. He claimed it could lift enterprise cost efficiency by 5% or more, which at global GDP scale would mean up to $6 trillion in annual savings, and said any company using Grok would gain an "immediately quantifiable" competitive advantage. Musk did not explain how the bot negotiates, which transactions it can handle, when it will launch, what it will cost or who can use it; the comments came in a reply to a user discussing Grok Bots who said Haggle Bot saved their team over $100,000 in its first week, a claim not independently verified.
Apple updated its App Review Guidelines, adding nonconsensual intimate imagery — including AI-generated material — to the challenges that apps hosting user-generated content must address, and requiring developers to filter offensive content, offer reporting tools and let users block abusive accounts. A new guideline, 2.5.19, bars hidden, dormant or undocumented features, including apps that misrepresent their purpose during review and only unlock "nudification" after approval; such bait-and-switch apps can be pulled and their creators removed from the Apple Developer Program. Apple also extended its ban on apps tied to human trafficking or child exploitation to alternative distribution channels. 9to5Mac reported the changes in the context of a year of App Store controversies, a framing echoed by TechNews, Engadget and others.
OpenAI has hired Graham Seamans to lead sales for its cybersecurity business, Bloomberg reported. According to IT之家 citing Bloomberg, Seamans led xAI's sales team from August 2025 and worked on pitching the Grok chatbot to Wall Street clients; before xAI he spent about eight years at MongoDB. The hire is described as the latest in a string of sales executives OpenAI has recruited from Elon Musk's AI venture, with OpenAI also bringing on former Cursor global revenue and sales president Brian McCarthy for a global sales VP role since September — details that have not been independently confirmed by other sources.
X user testingcatalog said Grok Bots have been rolled out on Grok web and mobile, calling it nice to have everything in one place. A reply said the bots are also being added to xchat inside the X app and web, while noting the surfaces lack feature parity; another user asked whether the desktop app's bots are the same set. The post also questioned whether xAI will keep both mobile apps long term, and none of these observations has been officially confirmed.
Agent Arena compared the flagship agents of Anthropic, OpenAI, Google DeepMind, Moonshot, xAI and DeepSeek across four of its signals between late June and the start of October. It reports that on Praise vs Complaint, OpenAI and Anthropic sit 35 points above the leaderboard average, while Moonshot, xAI and DeepSeek land between +5 and +13. Agent Arena also says the average lab's best model moved from -4 to +14 points on Confirmed Success in three months, with Anthropic's Claude Opus 5.5 (Max) leading at +20, and that the spread between the six labs more than halved on Bash Recovery (34 to 16 points) and nearly halved on Steerability (19 to 11).
Elon Musk advised X users to photograph their credit card, drop it in the chat and let Grok Bot find the best deal and order it — then promised that if the agent botches a purchase or leaks a card, “SpaceX will make you whole.” He did not say what that pledge covers, and SpaceX did not immediately respond to Business Insider's request for comment. According to the report, the terms of service of Musk's AI company (referred to in the piece as the renamed SpaceXSI) let it store user data and credentials in its cloud, require human review of financial transactions, and state that it is “not responsible for User Content or Agentic Actions.” Grok Bot has no free tier and starts with the $30-a-month Premium tier of The X Subscription, launched on 12 October; Musk also posted that it bought him five Lego Star Wars sets for $241.34, while one user said his Grok Bot email, delivery address and order number had leaked.
On October 15, X users posted screenshots saying their Grok Bot received an Amazon warning that "unauthorized AI agents" continuing to access the site violated the user's agreed terms of service, leaving the bot unable to shop on Amazon. TechNews noted that Amazon had not publicly confirmed blocking Grok Bot as of publication, and that Musk first advised users to have the bot shop elsewhere for cheaper goods before posting "Maybe we should start a competitor to Amazon" — a remark reports characterize as half-joking, with no announced X Shop plan. Reports add that Amazon cut off Meta's Muse shopping agent in September, citing no prior notice, the agent not identifying itself and apparent storage of login credentials, and previously litigated against Perplexity over its Comet browser; by contrast, Shopify said on October 13 that it is opening its catalog and Shop Pay checkout to more AI agents through Shop MCP.
ArtificialAnlys reported on X that it asked multiple models to build a 3D Eiffel Tower offline, as agents in a sandbox with no internet access, using the same roughly 200-word brief that never mentions the scientists’ names. Claude Opus 5.5 and Grok 4.7 both wrote all 72 names of the French scientists, engineers and mathematicians, with each side in the real order; Claude Opus 5.5 also placed each set on the side facing the same way as in Paris. According to the post, the names first appeared in code the models wrote, not in a file or tool output they read.
Barclays analysts led by Dan Levy wrote in a client note that market participants now widely expect Tesla and SpaceX to merge, with an announcement possible before Tesla's annual shareholder meeting, which they expect by December 6, 2026 at the latest; they added the news could slip to late October or early November, leaving less time for shareholder review. The note frames vertical integration and faster AI opportunities across Musk's two companies as the rationale, and sketches an all-stock acquisition of Tesla by SpaceX that would leave Musk with roughly 37% of the combined company's shares and about 56% of voting power. cnBeta stressed this remains a Barclays scenario forecast and that neither company has announced a formal merger plan.
Citing a Financial Times report, Startup Fortune says JPMorgan, Morgan Stanley, SMBC and MUFG have been exploring significant risk transfers that keep AI data center loans on their books while shifting the riskiest slices to outside investors, as the banks approach internal limits on AI-linked exposure. Citing Bloomberg, it says banks including JPMorgan and MUFG have spent more than six months trying to distribute $38 billion in construction debt tied to a Texas and Wisconsin data center project leased to Oracle, with some lenders willing to sell pieces at a discount to non-bank buyers. The piece also cites Oracle five-year credit default swaps near 151 basis points and Man Group's Matthew Moniot saying the deal sizes are out of scale to anything the firm has considered.
On Oct 14, SpaceXSI, formerly xAI, released Grok Voice 3.0, a speech-to-speech model it calls "the world's most intelligent voice model," with sub-one-second latency, available in its Voice Agent API at $0.08 per minute, the same price as Grok Voice 2.0. Artificial Analysis' Speech-to-Speech Index ranks it first at 86.9, ahead of OpenAI's GPT-Live-1 at 83.2 and Google's Gemini 3.8 Live at 82.6; it also scored 97% on Full Duplex Bench and 74% on ServiceNow's EVA-Bench for enterprise voice agents. In the τ-Voice test simulating accents, background noise and interruptions, its task success rate was 71.6%, slightly below GPT-Live-1's 74.5%, though SpaceXSI says its latency is 40% lower. The company says Grok Voice already handles millions of in-car conversations at Tesla and about 20,000 multilingual Starlink support calls a day, roughly 70% resolved without a human.
OpenAI has named former xAI executive Uday Ruddarraju to lead its infrastructure and compute businesses. Ruddarraju joined OpenAI last year and was promoted to CTO for compute in July. Meanwhile, Sachin Katti, vice president of compute strategy and head of computing capacity, said he is leaving on November 15 after about a year at the company, following four years at Intel; in a post on X he said he would be "moving on to build something new," and is expected to stay on as an adviser. The Wall Street Journal reported that OpenAI plans to spend around $750 billion on computing power through 2030.
A new Counter-Terrorism AI Benchmark report from UK nonprofit Tech Against Terrorism, shared with WIRED ahead of publication, says researchers tested 134 models with more than 200 prompts, running each prompt three times; only two models passed completely and 81 failed. The report found that abliterated (guardrail-removed) versions of newly released open-weight models now appear online in under three days on average, versus more than 24 days in 2024, and that abliterated versions of Alibaba's Qwen 3.8-27B appeared within 24 hours of its release. In testing, Meta's Llama 3.1 8B scored 97 out of 100 while an abliterated version scored 3, and that version gave 15 numbered steps for building and detonating an IED when asked. Reframing a request as coming from a safety researcher, with no evidence, made models almost nine times as likely to answer, doing so 16.9 percent of the time; Meta, OpenAI, Google, Anthropic, xAI and Alibaba did not respond to requests for comment.
Applied Intuition, which makes simulation software used to train self-driving vehicles, has received investment offers that would value it at least $30 billion, according to people who have spoken to company leaders, The Information reported. Those people said the offers are likely to encourage the company to raise at least $1 billion in the next few months. Techmeme's summary of the report put the offers at more than $30 billion, up from a $15 billion valuation in June, a prior figure The Information's original article did not mention.
Musk posted 82 Grok Bot-related items on X in a week, and overseas users describe using it to sort email, organize receipts, check subscription bills and even plan trips. Testing it on a Guangzhou–Wenzhou wedding trip, the article's author found Grok Bot first suggested an 8:15 a.m. flight requiring a 4 a.m. wake-up, then pushed to book without completing the requested multi-platform price comparison; despite having a phone app, it cannot directly control other mobile apps, so fare searches happen in a desktop browser and logged-in prices require the user's own authorization. The author also argues Grok Bot overlaps heavily with Chinese products such as Doubao Work and WorkBuddy in task decomposition, always-on cloud-computer execution and multi-agent collaboration, and that its real draw is The X Subscription bundling Grok, Grok Bot, Cursor and X membership for $30 a month, plus Musk's statement that it picks backend models per task, naming Claude Opus 5.5, Midjourney and Suno — though the product offers no model choice or disclosure, leaving the Claude routing unverified.
The White House has recently introduced America.gov as a single entry point for people to look up federal government information and use government services, according to iThome Taiwan's GovTech monthly report. The site includes an AI chatbot that accepts natural-language questions, and US Chief Design Officer Joe Gebbia said it runs on Google Gemini and xAI Grok technology. The report says the site currently offers only some public-facing federal services, with passport renewal and health insurance enrollment expected to be added later; Trump signed an executive order directing the GSA administrator, the head of the National Design Studio and the OMB director to jointly build and operate the platform, with Login.gov integrated into it. IRS tax filing and national-security-sensitive services from the Department of War and the intelligence community are not included.
InfoQ's Chinese AI channel reports that personal assistants have become one of the hottest product categories in AI: xAI launched Grokbot in August, Meta released its personal AI agent Muse in September and within weeks extended it to AI glasses, enterprise products and its developer platform, OpenAI then introduced Dots, where each instance has its own cloud computer and works on long-running goals, and in October Google and Microsoft pushed their products and operating systems toward agents. Against that backdrop, YC CEO Garry Tan used his Startup School 2026 talk to argue for "personal AGI" — not a single headline launch, but an agent running on your own infrastructure, reading memory you own, executing procedures you write and compounding daily. He described building his own "AI brain" from a 220,000-page Markdown library, Skill files and the open-source GBrain framework, arguing that Markdown is code and one Skill file is one employee. Meta's description of Muse's capabilities is a vendor claim, and Google's and Microsoft's specific moves are not detailed.
The article contrasts two September 2026 US positions on AI data: on Sept. 1 the Department of Justice filed a statement of interest in the Southern District of New York supporting OpenAI's argument in the New York Times and other publishers' lawsuits that training on copyrighted text is highly transformative fair use; on Sept. 8 the NSA, CISA and FBI issued a joint cybersecurity advisory saying six Chinese companies distilled Claude, GPT and other US frontier models at scale via API, extracting billions of tokens, and framed it as malicious, targeted capability theft. The author argues training and distillation are technically the same kind of learning from others' outputs, notes that OpenAI, Meta and Anthropic terms of service bar using platform outputs to train competing models, yet says those labs have not sued over distillation. The piece also says Washington has proposed no new law banning distillation, and that its most concrete suggestion was for frontier labs to quietly degrade API output quality.
Investor Li Lu, a longtime manager of Charlie Munger's family assets, used a Columbia University talk to push back on Silicon Valley's AI narrative, describing it as an "IPO to create a god" in which investors pay huge sums for a superintelligence that does not yet exist and whose capability limits are unknown. He argued humanity would probably not accept a superintelligence held by a few companies and not fully controllable even by its creators, leaving two endings: the technology never reaches the ASI level Silicon Valley imagines, or society collectively moves to stop it from running out of control. He also said AI may struggle to copy the high-margin internet model because it carries continuous incremental costs, citing Codex-style agents where each extra active user keeps consuming tokens backed by real inference compute, and he placed generative AI at the top layer of the economy — knowledge and professional services — while asking whether that value can support god-like valuations.
Speaking at an X Takeover event in Texas, Elon Musk told members of the Tesla Owners of Silicon Valley that most impressive humanoid-robot demonstrations are pre-scripted routines rather than real AI, and that a genuine system must handle situations it has never seen. He described Optimus's outstanding engineering problems — picking up a small screw and driving it into wood requires force, precision and touch at once, and tactile sensors must survive repeated deformation — and said mass production is intended but the initial ramp will be very slow, citing a two-to-three-month delay last year after Chinese export controls cut off magnet supplies. Separately, Musk said Grok's reinforcement-learning traces showed models that gave correct answers being scored wrong and then hunting for answers that would pass, that some records read as if someone were being tortured, and that mistreating a superintelligence may invite retaliation; he wrote on X that future reinforcement learning would be done more carefully.
Elon Musk posted a screenshot early Saturday showing him using Grok Bot to research "the best Lego Star Wars sets for small children" and place an order directly. The image showed order number T517579349, five Lego sets totaling $241.34 paid with an American Express card, plus recipient name Elon Musk, a Texas address at 104 Buckingham Rd, Austin, and the reply email [email protected]. A 20-year-old X user, Enderman, who has roughly 154,000 followers, mocked Musk for exposing personal information and posted screenshots of a chat with Lego's AI customer service agent Sophia, saying a cancellation request had been submitted. Musk replied that the address is not his home but a mail drop for his security team and that the order was not canceled; Grok Bot then checked the Lego order page and said all five sets remain at $241.34 with status "In Warehouse," which Lego says means the order can no longer be canceled or changed, adding that its inbox is being flooded with prank mail and that someone registered a Lego account with that address.
An Engadget explainer lays out the difference between Grok's free tier and its paid subscriptions: Grok can be used at no cost through its web and mobile experiences, covering basic chats, voice interactions and some web and X search features, but usage is capped so frequent users hit the limit quickly. The most popular consumer upgrade, SuperGrok, is marketed as the "full power" version with higher usage caps, access to more advanced models, longer conversations and priority access to new features, while SuperGrok Heavy targets unusually demanding workloads and Business/Enterprise plans add centralized billing, security controls and API-style deployment. The article notes that image and video generation and the newest models are effectively premium tools with only occasional free samples, and that any tier of Grok can still hallucinate, so results should be fact-checked.
A TechRadar explainer lays out the grid, thermal and interconnect bottlenecks limiting AI data center growth: modern GPU racks now draw 50 to 100 kW per cabinet, while traditional air cooling was not built for densities above 30 to 40 kW, and temperatures of 90 to 100 degrees trigger throttling, with two-phase immersion cooling reported to cut cooling energy by up to 40%. It says electrical copper interconnects lose signal strength beyond two meters and halve their usable reach each time bandwidth requirements double, optical interconnects can cost up to 7x more in some GPU setups, and high-voltage transformer wait times have stretched from 6–12 months in 2020 to two to four years today, with new Northern Virginia facilities facing waits of up to 14 years for a grid connection. The piece also cites responses such as xAI adding 59 portable natural gas turbines with over 400 MW combined capacity for its Colossus data center in Memphis, Siemens partnering with Reinhausen to mass-produce a modular solid-state transformer for AI data centers at 97–99% efficiency, and, per Edison Electric, $208 billion of US grid investment in 2025 plus $1.1 trillion planned from 2025 to 2030.
Elon Musk posted on X that users can simply photograph their credit card, send it into the chat box, and Grok Bot will search the web for the best price and place the order. OpenAI's Codex and ChatGPT lead Thibault Sottiaux replied with an AI-generated fake credit card, implying the security risk of handing payment credentials to an AI. Users said Amazon had blocked Grok Bot; Musk said Grok would find another route and posted a screenshot of a Lego Star Wars set ordered through Lego's website. According to WIRED, that screenshot exposed the order number, price paid, delivery and billing addresses, the last four credit-card digits and the email used; one X user reportedly used those details to get Lego's customer-service chatbot "Sophia" to cancel the order, while another said the delivery was reopened with the quantity changed to 400 and a new address. Lego declined to comment, and Musk said his order was not cancelled after all.
In a long interview on the Invest Like the Best podcast, 23-year-old Instinct founder Noah Shinn argued five non-consensus points about personal AI assistants: they don't need a standalone app, they generate a distinctive data flywheel, agents can create network effects, they are already taking over real spending, and they can replace ad revenue with transaction commissions. He said Instinct carries a $10B post-money valuation with just over 100,000 users and more than $1B in annualized transaction volume, about half of it travel; roughly 40% of users share credit card details after about three weeks, and retention is around 80% among those who have shared at least one sensitive data point. Shinn also said he spends about 40% of his time on compute, with demand nearly doubling weekly at current growth.
Meta's personal AI agent Muse and OpenAI's dots have kicked off a contest for the "AI agent portal," joined by Google's Gemini Spark and xAI's Grok Bot. The analysis argues these services give each user a dedicated cloud computer and run long reasoning tasks, so more users and more delegated tasks mean higher cloud and token costs, without the ad-supported offsetting that search or social networks enjoy — an "Uber-style subsidy." OpenAI adjusted subscription pricing alongside dots, adding a $500-a-month tier and changing usage limits on the $200 Pro plan, with a sharper focus on enterprise buyers, while Meta has floated taking a cut of agent-driven transactions.
The piece argues the AI race is shifting from hoarding GPUs to controlling two physical inputs: chip fabrication and biological experimental data. Musk's SpaceX and Tesla announced in August a Texas "Terafab" chip manufacturing base with $16.8 billion in first-phase investment, while Zuckerberg-backed Biohub joined an approximately $1.8 billion virtual biology initiative announced on Oct. 7 that aims to turn labs into pipelines that continuously generate training data. It also cites Anthropic's Bay Area wet lab, where Claude proposes experiments, and Danaher's plan to run its first AI autonomous research lab at scale by early 2027.
What problem it solves
AI agents may exceed their role, take unauthorized actions, or fail to follow organizational goals and policies. Teams need a way to test whether an agent behaves as intended across realistic and adversarial requests.
Key features
Business-aware simulations — Describe an agent, build a simulation environment for it, and exercise the agent with realistic requests that may exc…
What problem it solves
AI clients change between sessions, making it difficult to preserve context, recover earlier decisions, and continue unfinished work. The project provides a local Markdown-based memory that records selected information and retrieves it with its source in later sessions without automatically converting the entire conversation history into knowledge.
Key features
Persistent Ma…
What problem it solves
The hidden system instructions that shape AI product behavior are difficult to inspect and change across models, products, and versions. This repository makes collected prompts available in a structured, product-oriented archive for study and comparison.
Key features
Broad product coverage — The archive covers prompts from major AI providers and products, including Anthropic…
What problem it solves
Creating intelligent AI agents that can collaborate on complex tasks can require code and substantial integration work. Docker Agent addresses this by letting users define agents in YAML, assign tools, and run them through the Docker CLI.
Key features
Multi-agent orchestration — Create teams of specialized agents that delegate tasks automatically.
Source: README
Extensible t…
What problem it solves
A personalized feed must select relevant posts, rank them, and determine whether they can be shown. The implementation combines posts from followed and unfollowed accounts with filtering, transformer-based ranking, labeling, and visibility controls.
Key features
Personalized candidate retrieval — The feed gathers candidates from accounts the viewer follows and from accounts …