Skip to content

All news

12 today

Apr 24Friday

QbitAI · WeChat

Claude admits three issues: downgraded reasoning, cleared memory, and constrained output

Anthropic said on April 23 that three Claude issues hurt quality: Claude Code default reasoning was changed from high to medium on March 4 while the UI still showed high. A March 26 cache bug cleared thinking state every turn for 15 days, and an April 16 prompt limit of 25 words between tool calls and 100 words in final replies cut Opus 4.6/4.7 by 3% before a rollback four days later.

Why it matters: Anthropic publicly reviewed three recent Claude quality problems, including a default reasoning effort lowered from high to medium while still displaying high, a gap between what users pay for and what they get.

Bloomberg Technology

AI Coding Firm Cognition in Funding Talks at $25 Billion Value

Cognition is in early talks to raise funding at a $25 billion valuation, more than double its prior valuation. The RSS snippet says demand for AI software-development firms is rising, but the post does not disclose investors, round size, or timing.

Why it matters: Cognition is negotiating a new round at a reported $25 billion valuation, more than double its last round, a pace that shows continued market bets on AI coding.

Bloomberg Technology

Bezos’ Physical AI Lab Has Closed Round at $38 Billion Value

Project Prometheus closed a $10 billion round at roughly a $38 billion valuation, led by Jeff Bezos with former Google executive Vik Bajaj. The RSS snippet discloses only the round size, valuation, and principals; the post does not disclose investors, product scope, or timing. The key signal is the scale: a single $10 billion round prices a physical AI lab at $38 billion.

Why it matters: Bezos personally led Project Prometheus's $10 billion round at a roughly $38 billion valuation, giving a concrete reference for capital going into physical AI.

Financial Times · Technology

UK in talks with Anthropic over Mythos access for banks

The UK is in talks with Anthropic about bank access to Mythos, with the stated use case being cyber security testing. The RSS snippet says British lenders are seeking advice from US groups testing the model; the post does not disclose Mythos capabilities, deployment terms, customer scope, or timing. The key issue is whether finance gets controlled access to a frontier offensive-defensive security model, not a generic AI rollout.

Why it matters: The UK government is talking with Anthropic about giving domestic banks access to the Mythos model for cybersecurity testing, a controlled path for opening a highly sensitive security tool.

The Verge · AI

Meta is laying off 10 percent of its staff

Meta plans to cut about 10 percent of staff in May, affecting roughly 8,000 employees, according to a memo cited by Bloomberg. Meta will also close about 6,000 open roles; the cuts follow heavier AI spending, with 2026 capex guided to $115 billion to $135 billion versus $72.22 billion in 2025.

Why it matters: Meta plans to cut about 8,000 jobs in May and close 6,000 open roles, while its 2026 AI capex guidance nearly doubles from 2025.

Hacker News front page

Meta to cut 10% of jobs, or 8,000 employees

Meta plans to cut 10% of its workforce, or 8,000 employees, and not hire for 6,000 open roles. A Bloomberg-cited internal memo says the cuts start May 20; Meta had not responded to TechCrunch for comment. The key signal is capital reallocation: the memo ties the cuts to efficiency and offsetting AI and other investments.

Why it matters: Meta will cut 8,000 jobs from May 20 and freeze 6,000 open roles, with its chief people officer saying in an internal memo the move frees money for AI and other areas.

Hacker News front page

An update on recent Claude Code quality reports

Anthropic said three product-layer changes degraded Claude Code quality for Sonnet 4.6, Opus 4.6, and Opus 4.7, while the API was unaffected; all were fixed on April 20 in v2.1.116. The changes were lowering default reasoning effort on March 4, a March 26 bug that cleared prior thinking every turn after sessions sat idle for over an hour, and an April 16 prompt tweak to reduce verbosity that hurt coding quality. The signal for practitioners is sharp: product and prompt changes can degrade code performance even when model and inference evals do not reproduce it early.

Why it matters: Anthropic traced a Claude Code quality drop to three product-layer changes, not the API or model, showing non-model tweaks can shape the coding experience.

Financial Times · Technology

Microsoft to offer 7% of US staff voluntary redundancy for the first time

Microsoft will offer voluntary redundancy to 7% of its US staff for the first time. The RSS snippet says long-serving employees can take buyouts, while the company plans to spend $140bn on AI this year; the post does not disclose affected teams, roles, or timing. The signal to watch is AI capex expansion paired with workforce reshaping in the same year.

Why it matters: Microsoft opened voluntary redundancy to about 7% of US staff while planning $140 billion in AI spending this year, a look at where the money goes.

Apr 23Thursday

Financial Times · Technology

DeepSeek targets a $20bn valuation to stop poaching of staff

DeepSeek is seeking its first funding round at a $20bn valuation to reduce rival poaching of researchers. The RSS snippet discloses prior defections and that this is its first raise, but the post does not disclose round size, investors, or headcount lost. The real signal is talent retention, not the headline valuation.

Why it matters: DeepSeek's first outside round, at a reported $20bn target valuation, puts a price on the equity it needs to keep researchers from being poached.

Xinzhiyuan · WeChat

Tashi Zhihang raises $455.0 million in a Pre-A round, with Sequoia China and Hillhouse jointly leading

Tashi Zhihang said on April 16 it closed a $455.0 million Pre-A round led by Sequoia China, Hillhouse Ventures, and Meituan, which the post says set China records for embodied AI single-round and Pre-A financing. The post also says its AWE3.0 four-modal model lifted unseen-view task success by 3x and cut execution jitter by about 45%, and that its A1 robot set a Guinness record in sub-millimeter wire-harness assembly within one hour. What matters is whether model, data, and deployment keep reproducing; the post does not disclose valuation or deal terms.

Why it matters: Tashi Zhihang's $455m Pre-A, led by Sequoia China, Hillhouse and Meituan, sets a domestic record for a single embodied-AI round and comes with AWE3.0 and A1 deployment figures.

New York Times Chinese

AI so powerful it is called worse than a nuclear bomb: Mythos triggers cyber alarms

Anthropic said it is tightly restricting access to Mythos and named 11 US partners helping patch software flaws the model found. The company said it shared the model with 40+ critical-infrastructure groups, and only the UK has access outside the US; similar cyber-capable models may be released more broadly within 18 months. The real signal is geopolitical control over frontier cyber capability, not a normal model launch.

Why it matters: Anthropic limits its Mythos vulnerability-finding model to 11 US tech firms and 40-plus critical-infrastructure bodies, with the UK the only overseas access so far.

Bloomberg Technology

SoftBank Seeks $10 Billion Loan Backed by OpenAI Shares

SoftBank is seeking a $10 billion loan backed by its OpenAI shares. The RSS snippet says the move adds debt to support its AI push; the post does not disclose tenor, rate, collateral ratio, or use of proceeds. The key signal is margin financing, not a generic AI bet.

Why it matters: SoftBank is negotiating a $10 billion margin loan secured by its OpenAI stake, a case study in leveraging existing AI equity.

Bloomberg Technology

Microsoft Commits $18 Billion to Build Australian AI Capacity

Microsoft said it will spend A$25 billion ($17.9 billion) in Australia by the end of 2029, its largest investment there. The post discloses only the total, location, and deadline; it does not disclose data center scale, GPU count, customer allocation, or product scope.

Why it matters: Microsoft plans to spend A$25 billion on Australian AI infrastructure by the end of 2029, its largest investment in the country.

Bloomberg Technology

Core Scientific Raises $3.3 Billion From AI Junk-Bond Offering

Core Scientific raised $3.3 billion through a high-yield note sale tied to AI infrastructure construction. The post discloses the financing size, debt type, and broad use case, but not the coupon, maturity, buyers, or specific projects. The key issue is funding cost versus cash flow, not any disclosed compute deployment detail.

Why it matters: Core Scientific's $3.3 billion junk-bond raise for AI infrastructure shows how bitcoin miners are financing the shift to AI compute.

TechCrunch · AI

How SpaceX preempted Cursor's $2B fundraise with a $60B buyout offer

Cursor was on track to close a $2B funding round this week, but halted talks after SpaceX offered a $10B “collaboration fee” and a path to a $60B acquisition. The RSS snippet discloses the two figures and timing only; the post does not disclose deal structure, timeline, or responses. The key point is that a buyout path preempted a primary fundraise.

Why it matters: SpaceX stepped in with a $1 billion partnership fee plus a $6 billion acquisition option, and Cursor halted a $2 billion raise at a $50 billion valuation.

The Verge · AI

Anthropic’s Mythos rollout left out America’s cybersecurity agency CISA

Axios reported that Anthropic’s vulnerability-finding model Mythos Preview is already in use at multiple US federal agencies, but CISA still lacks access. The snippet names the Commerce Department and NSA as users, and says the Trump administration is negotiating broader access; the post does not disclose model specs, pricing, or why CISA was excluded. The signal is governance, not just product rollout.

Why it matters: Anthropic's Mythos Preview reached Commerce and the NSA, but CISA, which runs national cyber defense, was left out, showing how uneven federal AI access is.

Apr 22Wednesday

TechCrunch · AI

Exclusive: Google deepens Thinking Machines Lab ties with new multibillion-dollar deal

Thinking Machines Lab signed a multibillion-dollar deal with Google Cloud for AI infrastructure powered by Nvidia’s latest GB300 chips. The snippet discloses the deal size, cloud provider, and chip generation; the post does not disclose term length, compute volume, delivery timeline, or workload details. The real signal is GB300 entering a top lab’s procurement stack, not just launch-stage specs.

Why it matters: Mira Murati's Thinking Machines Lab signed a multi-billion-dollar, multi-year compute deal with Google Cloud using Nvidia GB300 chips for training and inference, adding one more data point on how top labs buy silicon.

Hacker News front page

Kernel code removals driven by LLM-created security reports

Linux kernel maintainers are proposing to remove several legacy networking components to reduce the workload from rising LLM-generated security reports. The post names ISA and PCMCIA Ethernet drivers, two PCI drivers, the ax25/amateur-radio subsystem, ATM, and ISDN; one patch says hamradio code has long been a bug and syzbot magnet, with no one stepping up to handle the AI-report influx. The real issue is not LLMs helping cleanup, but unmaintained code collapsing under report volume.

Why it matters: A Linux kernel maintainer proposed removing unmaintained network components such as ISA, PCMCIA Ethernet drivers, ax25, ATM and ISDN, after a flood of LLM-generated security reports left no one to verify them.

Bloomberg Technology

Tencent, Alibaba in Talks to Join DeepSeek’s First Funding Round

Tencent and Alibaba are in talks to join DeepSeek’s first funding round, and the snippet confirms this is DeepSeek’s maiden financing. The RSS text discloses only the talks and the first-round status; it does not disclose the round size, valuation, lead investor, or timing. What matters is whether strategic capital from two Chinese internet giants also brings compute or distribution terms, but the post does not disclose them.

Why it matters: Two of China's biggest cloud and model players talking to the same startup would reshape who backs the country's leading open-weight lab.

The Verge · AI

Anthropic’s most dangerous AI model just fell into the wrong hands

Anthropic’s Claude Mythos Preview was accessed by a small group of unauthorized users through a contractor’s access plus common internet sleuthing tools. The snippet says the model can identify and exploit flaws in major operating systems and browsers; the post does not disclose the group size, dwell time, or remediation status. The key issue is access control failure, not the headline’s danger framing.

Why it matters: A contractor permission slip let Discord users reach Anthropic's internal Claude Mythos Preview for two weeks, showing how thin third-party access control is at frontier labs.

Hacker News front page

Meta staff protest surveillance software on work PCs

Meta reportedly told staff to soon run a tool called Model Capability Initiative on work PCs to record keystrokes, prompting employee protest. The visible text discloses the tool name, and the Reuters link points to mouse-movement and keystroke capture; the post does not fully disclose scope, rollout timing, or opt-out terms. The key issue is whether Meta is routing internal behavior data into AI capability building.

Why it matters: Meta plans to run software on work PCs that logs keystrokes and mouse movement to train AI models, and staff are unhappy about it.

Financial Times · Technology

OpenAI in talks to commit up to $1.5bn to private equity joint venture

OpenAI is in talks to commit up to $1.5bn to a private equity joint venture. The RSS snippet says the new company is meant to help deploy AI in businesses owned by PE firms; the post does not disclose the partner, deal structure, or timeline. This is not a model launch but a distribution bet on enterprise deployment.

Why it matters: OpenAI is in talks to commit up to $1.5bn to a private equity joint venture, a capital route to deploy its models inside the traditional companies those funds own.

Financial Times · Technology

Insurers move to cap cyber payouts related to AI and 'LLMjacking'

Beazley and QBE are proposing caps on cyber insurance payouts tied to AI and 'LLMjacking'. The RSS snippet discloses only that these groups want limits; the post does not disclose cap size, trigger conditions, or timing. The key issue is how policy wording defines AI-linked losses.

Why it matters: Beazley and QBE are capping cyber insurance payouts tied to AI and LLMjacking losses, so how policies define AI-related loss decides what companies can recover.

Xinzhiyuan · WeChat

Musk sets Cursor deal terms: SpaceX can buy it for $60B or pay a $10B collaboration fee

SpaceX disclosed terms to work with Cursor: it can acquire the startup this year for $60B or pay a $10B collaboration fee. The post says xAI already supplies Colossus compute to train Cursor's Composer, while Cursor had over $2B annualized revenue and 1M+ daily users by Feb. 2026. The key point is the bundle: SpaceX gets a coding product, and Cursor gets model and compute support.

Why it matters: SpaceX gave Cursor a take-it-or-leave-it choice this year, and the two price tags shape how the AI coding race gets financed.

Bloomberg Technology

Japan Finance Minister to Meet Banks to Discuss Anthropic Mythos Threat

Japan Finance Minister Satsuki Katayama plans to meet the country’s biggest banks as early as this week to discuss threats tied to Anthropic’s latest AI model, Mythos. The RSS snippet confirms large banks and other financial institutions are included; the post does not disclose Mythos’s capabilities, the risk type, or any regulatory action. The real signal is that Japan may be moving frontier-model risk into formal banking discussions.

Why it matters: Japan's finance minister plans to meet major banks this week over risks Anthropic's new Mythos model poses to finance, putting frontier model risk on the bank supervision agenda.

Bloomberg Technology

RBA Is Monitoring Anthropic's Mythos AI Over Cyberattack Fears

The Reserve Bank of Australia is monitoring Anthropic's Mythos AI after the model was described as capable of sophisticated cyberattacks. The Bloomberg RSS snippet says Anthropic made that claim; the post does not disclose scope, technical details, or timeline.

Why it matters: Australia's central bank is watching Anthropic's Mythos after Anthropic itself said the model can carry out complex cyberattacks, a concrete case of a central bank tracking model security.

Financial Times · Technology

SpaceX obtains right to buy AI start-up Cursor for $60bn

The headline says SpaceX obtained the right to buy AI start-up Cursor for $60bn. The RSS snippet adds one fact: Elon Musk’s rocket and AI group is trying to catch up with OpenAI and Anthropic; the post does not disclose the trigger, timeline, deal structure, or regulatory terms. The key issue is whether this is tied to future financing or control rights, not a completed acquisition.

Why it matters: SpaceX holds the right to buy AI coding tool Cursor for $60bn, a figure that frames how code assistants are valued as Musk's companies chase OpenAI and Anthropic.

Financial Times · Technology

Anthropic investigating unauthorised access to powerful Mythos AI model

Anthropic is investigating unauthorised access to its Mythos AI model. The RSS snippet says it limited the new tool’s release over concerns about hacking ability. What matters is the breach scope and release status; the post does not disclose impacted accounts, capability limits, or timeline.

Why it matters: Anthropic is investigating unauthorized access to its Mythos model, which it had already restricted over hacking concerns, raising questions about capability control and access security.

TechCrunch · AI

SpaceX is working with Cursor and has an option to buy the startup for $60B

SpaceX is working with Cursor and holds an option to acquire the startup for $60B. The RSS snippet discloses the collaboration and purchase option, but not the term, trigger conditions, ownership impact, or payment mix. The sharper signal is strategic weakness: the snippet says neither Cursor nor xAI has proprietary models matching leading offerings from Anthropic and OpenAI.

Why it matters: SpaceX is building coding AI with Cursor and holds an option to buy the startup for $60B later this year, tying the two companies' tooling and engineering work together.

The Verge · AI

SpaceX cuts a deal to maybe buy Cursor for $60 billion

SpaceX announced an either-or deal: buy AI coding platform Cursor for $60 billion or pay a $10 billion fee. The RSS snippet says this could help xAI's coding tools chase Anthropic; the post does not disclose the structure, timing, or IPO linkage. Watch the $10 billion breakup fee, not just the tentative acquisition headline.

Why it matters: SpaceX signed a take-it-or-leave-it structure with Cursor at a $60 billion purchase price and a $10 billion breakup fee if it walks away, with neither the deal terms nor a timeline disclosed.

Bloomberg Technology

SpaceX Has Deal for Right to Acquire Cursor for $60 Billion

SpaceX said it signed a deal giving it the right to acquire AI coding startup Cursor later this year for $60 billion. If it does not proceed, it can pay $10 billion for the companies' collaboration; the post does not disclose trigger terms, scope, or regulatory details.

Why it matters: A rare structure: SpaceX buys the right to acquire Cursor for $60 billion, or pays $10 billion to keep the partnership if it walks away.

Bloomberg Technology

Anthropic’s Mythos Model Is Being Accessed by Unauthorized Users

A small group of unauthorized users accessed Anthropic’s new Mythos model, Bloomberg reported, citing a person familiar with the matter and reviewed documents. The snippet says Anthropic considers Mythos powerful enough to enable dangerous cyberattacks; the post does not disclose the user count, access path, time frame, or remediation. The real issue is access control failure, not a normal product launch.

Why it matters: A small number of unauthorized users accessed Anthropic's unreleased Mythos model, which the company judges capable enough to support dangerous cyberattacks, making this a story about access control rather than a product leak.

Bloomberg Technology

Apple CEO Transition: Hardware, AI, and China | Bloomberg Tech 4/21/2026

Apple said hardware chief John Ternus will replace Tim Cook as CEO on Sept. 1. Cook will become executive chairman, and Bloomberg says his corporate diplomacy and ties to Donald Trump will remain available to Apple. The key signal is hardware priority; the title mentions AI and China, but the post does not disclose specific plans.

Why it matters: Apple puts a hardware executive in the CEO seat on September 1, while keeping Tim Cook around for government relations.

Financial Times · Technology

Elite law firm Sullivan & Cromwell admits to AI 'hallucinations'

Sullivan & Cromwell apologized to a judge over AI-related errors in a bankruptcy case, and the title says the firm admitted to “hallucinations.” The RSS snippet discloses only that partners bill above $2,000 per hour and the errors were software-driven; the post does not disclose the AI tool, error count, or court response. Watch the process failure: premium human review still did not catch checkable mistakes.

Why it matters: A law firm billing over $2,000 an hour apologized to a judge after AI-generated content with errors was filed in a bankruptcy case.

The Verge · AI

AI backlash is coming for elections

An Ipsos poll found over 60% of both Republicans and Democrats support government regulation of AI and slower development. The RSS snippet also says US communities are resisting data center projects and anger at AI firms is rising online, but experts say AI is still not a central campaign issue. The post does not disclose sample size, timing, or specific election cases.

Why it matters: Ipsos polling shows over 60% of voters in both parties back government AI regulation and slower development, while communities resist data centers.

TechCrunch · AI

Report says Clarifai deleted 3 million photos OkCupid provided to train facial recognition AI

Clarifai deleted 3 million photos from OkCupid after an FTC settlement, and the images had been used to train facial recognition AI. The RSS snippet says the data sharing request dates to 2014 and OkCupid executives had invested in Clarifai. The post does not disclose the settlement terms, deletion verification, or model impact.

Why it matters: Clarifai's FTC settlement forced it to delete 3 million OkCupid user photos and the facial recognition model trained on them, data OkCupid handed over in 2014 while an executive was investing in Clarifai.

Apr 21Tuesday

The Verge · AI

John Ternus’ first big problem is AI

Apple said hardware chief John Ternus will become CEO on September 1, and the official release does not mention AI once. He has spent 25 years at Apple and is its first hardware-background CEO in about 30 years; the post does not disclose his AI strategy, Siri plans, or org changes. The real issue is Apple’s visible AI gap after last year’s WWDC criticism.

Why it matters: Apple named hardware veteran John Ternus as CEO starting September 1, an announcement that never mentions AI even as Apple's AI gap stays on the table.

Bloomberg Technology

Apple Bets New CEO John Ternus Will Bring Back Jobs-Era Decisiveness

Apple named John Ternus to succeed Tim Cook as CEO, according to a Bloomberg post published on April 21, 2026. The body is largely missing and only confirms succession plus AI and product challenges; the post does not disclose timing, org changes, or AI strategy details. Watch for decision rights and AI roadmap, not the Jobs-era framing.

Why it matters: Apple confirmed hardware engineering chief John Ternus as its next CEO, and the report flags the AI and product-line challenges he inherits.

New York Times Chinese

Who is John Ternus, Apple's incoming empire heir?

Apple appointed 50-year-old hardware chief John Ternus as its next CEO on April 20, succeeding 65-year-old Tim Cook. The post says Apple accelerated succession planning last year, and Cook told leaders he was tired and wanted a lighter workload; he may become board chair. For AI practitioners, the key point is Apple still trails in generative AI, while Ternus is described as a hardware and supply-chain operator, not a proven AI product pivot leader.

Why it matters: Apple named hardware engineering head John Ternus to succeed Cook as CEO, the company's first pure-hardware chief in 30 years, long responsible for iPhone, iPad and Mac hardware and supply chain.

New York Times Chinese

Tim Cook to Step Down as Apple CEO

Apple said 65-year-old Tim Cook will step down as CEO in September, with hardware chief John Ternus, 50, taking over and Cook becoming executive chair. During Cook’s nearly 15-year tenure, Apple’s annual profit rose to over $110 billion and its market value reached $4 trillion; the post also says Apple has stayed largely on the sidelines as rivals commit hundreds of billions to AI.

Why it matters: Cook moves to executive chairman in September and hardware chief Ternus becomes CEO, with Apple's profit and market value set against its thin pipeline of new products.