DX Today | No-Hype Podcast & News About AI & DX

Palantir's chief technology officer says most customers who set out to cut staff with AI failed to even achieve that.

Why Most AI Layoff Plans Fail

Palantir's chief technology officer says most customers who set out to cut headcount with AI failed to do it, while those chasing work they could never afford are winning. Chris and Laura weigh that claim against Goldman's data on displaced workers, a surprise jump in enterprise AI spending, and what Apple, Accenture and Revolut are doing about the bill.

Hosted by Chris and Laura.

The DX...

OpenAI just took in eleven billion dollars at a one point two five trillion dollar valuation, yet the round still falls short of the thirty billion it wanted.

DX Today AI Daily Brief - Saturday, November 7, 2026

OpenAI collects eleven billion dollars at a one point two five trillion dollar valuation, though the round is still short of its original target. Shield AI agrees to buy radar maker Echodyne for five point three billion dollars. Google takes over Crusoe's unfinished Wyoming data center project with a planned seventeen billion dollar investment. SoftBank and ABB look for a...

Google builds its own power plant, Uber counts four agent rides, and an AI model cracks the Linux kernel.

Executive Briefing: Google Wyoming, Uber Agents, AI Law, Kernel

Google commits seventeen billion dollars to Wyoming and a fuel cell power plant. Uber says AI agents booked just four rides in its busiest week. Democrats promise AI hearings and guardrails. OpenAI's cyber model builds root exploits for nine hundred forty eight Linux kernel flaws.

Hosted by Elena and Marcus.

The DX Today Executive Briefing: a deep dive review of every Tuesday and Friday...

House Democrats are weighing an AI select committee with subpoena power, and companies that relied on voluntary safety pledges may have to hand over their test results.

Democrats Win the House and Promise AI Guardrails

After the November midterms, House Democrats are promising AI hearings, a possible select committee with subpoena power, and bipartisan bills on child safety and liability. Brian and Jessica break down what is likely to change first for people who build and buy AI.

Hosted by Brian and Jessica.

The DX Today Podcast brings you daily deep dives...

OpenAI just shipped a model so capable at finding software flaws that it rates it Critical, its highest cyber risk level.

DX Today AI Daily Brief - Friday, November 6, 2026

OpenAI releases GPT-6 Cyber, a vulnerability hunting model it rates Critical, to vetted Daybreak Red customers. AI stocks lead Wall Street's worst session in more than a year as the S&P 500 falls 4.36 percent and Nvidia drops about 9 percent. Thinking Machines Lab launches Inkling 2 Small, a 594 billion parameter mixture of experts model, amid reports of a $5 billion raise. Shield AI agrees to buy radar maker Echodyne for...

Apple just told investors its AI agents may resist directions and take actions that cannot be undone, and the same day CrowdStrike started selling the brakes.

Apple and CrowdStrike Confront Rogue AI Agents, November 5, 2026

Apple's new annual report warns that AI agents may act outside their permissions, resist directions and take actions that cannot be undone, while CrowdStrike launched an AI Kill Switch in Falcon Guardian and agreed to buy Fiddler AI and Dreadnode for about $495 million. Chris and Laura unpack what it means when agents need brakes, and what businesses and everyday users should ask...

After voters rejected nearly every data center measure on the ballot, OpenAI says it will pay the full cost of powering the data centers it owns.

DX Today AI Daily Brief - Thursday, November 5, 2026

OpenAI pledges to pay the full cost of powering its own data centers and to drop nondisclosure agreements with officials after voters rejected nearly every data center measure on the ballot. China blocks SoftBank's $5.4 billion takeover of ABB's robotics division. Microsoft declines to expand its Texas computing deal with Nscale. Tencent launches Hy4, a 770 billion parameter model with computer use. Isomorphic Labs...

After months of letting traders and bankers build their own software with AI agents, Goldman Sachs just handed those tools back to its 12,000 engineers.

Goldman Sachs Pulls AI Coding Tools From Traders and Bankers, November 4, 2026

Goldman Sachs has restricted Claude Code and GitHub Copilot to its roughly 12,000 engineers after a productivity and risk review, reversing months of open access for traders and bankers. Brian and Jessica unpack why the bank drew the line, how Revolut is taking the opposite path with self hosted models, and what every company should learn about letting AI take actions instead...

AMD's data center sales jumped 97 percent in a single year, while Oracle walked away from a $7 billion chip deal with Tencent under pressure from Washington.

DX Today AI Daily Brief - Wednesday, November 4, 2026

AMD posts the largest and most profitable quarter in its history as data center revenue nearly doubles to a record. Oracle cancels a roughly $7 billion chip leasing deal with Tencent after pressure from the Trump administration, the Financial Times reports. An adviser to the EU's top court says Google's Gemini may infringe press publishers' rights when it reuses news articles. Canva announces a...

Even the best AI agent completes just twelve percent of expert work, and regulators say you own whatever your agent does.

Executive Briefing: ADNOC's AI Agents, GDPx 12% Ceiling, UK Agent Law

ADNOC becomes the first energy company to deploy OpenAI Frontier, aiming agents at field development plans that take months. SemiAnalysis launches GDPx and finds the best AI agent scores just 12% on expert professional work. Four UK regulators declare that existing law already governs personal AI agents, and the ICO sets a November 20 evidence deadline. The NCSC warns AI malware is highly likely to hit British...

Deep in the risk factors of its IPO filing, Anthropic says it may imminently hit its own threshold for AI that dramatically speeds up AI research.

Anthropic Warns Its AI May Soon Speed Up Its Own Research

In its amended IPO filing, Anthropic says it may imminently meet its own Responsible Scaling Policy threshold for dramatically accelerated progress attributable to AI, though no current model has been designated as crossing it. Chris and Laura unpack what the threshold means, what safeguards it triggers, why skeptics call it part marketing, and what it signals for business leaders...

Marvell just agreed to the largest deal ever between two public tech companies, betting $115 billion that light, not copper, will carry AI's data.

DX Today AI Daily Brief - Tuesday, November 3, 2026

Marvell agrees to buy Lumentum in a $115 billion all stock deal betting on optical networking for AI data centers. Anthropic's amended IPO filing shows nearly $17 billion in third quarter revenue, slowing sequential growth, and about $895 billion in compute and infrastructure commitments. Apple beats estimates with $114.5 billion in quarterly revenue but warns memory costs and its new AI powered Siri will pressure holiday margins. Palantir tops $2...

China added roughly ten times more new power capacity than the United States last year, and it is now wiring AI data centers straight into wind and solar at prices twenty percent below the grid.

China Plugs Its AI Data Centers Into Wind and Solar, November 2, 2026

China is wiring giant AI data centers directly into wind, solar and battery storage, led by a two gigawatt Envision Energy campus in Inner Mongolia that promises power twenty percent below grid prices. Meanwhile the United States is leaning on new gas plants as local opposition to data centers grows...

Abu Dhabi's state oil company just became the first energy firm anywhere to run OpenAI's agent platform across its operations.

DX Today AI Daily Brief - Monday, November 2, 2026

Abu Dhabi signs four agreements with OpenAI as ADNOC becomes the first energy company to deploy the OpenAI Frontier agent platform with G42. Saudi Aramco closes its $2.6 billion purchase of a 49 percent stake in Humain. South Korea proposes nearly doubling AI and chip spending to about $16 billion in its 2027 budget. Microsoft AI chief Mustafa Suleyman warns against building conscious AI. Tesla expands Robotaxi to Fort Worth. Affle agrees...

Meta could burn through about $132 billion in free cash flow over three years, and one rating agency just said it no longer believes the spending will slow down.

S&P Global Warns Meta Over Its AI Spending, November 1, 2026

S&P Global moved its outlook on Meta from stable to negative, warning that AI capital spending could top $225 billion next year, push free cash flow deeply negative, and lift lease driven leverage past its downgrade threshold by 2027. Rick Spair and Laura unpack the numbers, why Meta was singled out while cloud rivals were not, and what business...

OpenAI trimmed its fundraising target by ten billion dollars in a single week, while Apple quietly picked a Chinese AI model for iPhones in China.

DX Today AI Daily Brief - Sunday, November 1, 2026

OpenAI cuts its fundraising target to about $20 billion at a $1.25 trillion valuation as AI stocks slide, while Foxconn says there is no AI bubble. Apple taps Alibaba's Qwen to bring Apple Intelligence to iPhones in China. Saudi Aramco closes its purchase of a 49 percent stake in national AI company HUMAIN. DeepSeek open sources Super agent, a prototype autonomous coding agent aimed at Claude...

Google still crawls Reddit as hard as ever, yet Reddit says it lost more than a million search visits every week in October.

Reddit Blames Google AI Mode for Vanishing Search Traffic, October 31, 2026

Reddit says referrals from Google search fell by more than a million visits a week in October, and its chief executive blames Google's AI Overviews and AI Mode. Chris and Laura unpack why Reddit is moving to a primarily logged in experience, why Alphabet admits AI searches earn less than classic ads, and what every business that depends on search traffic should do...

OpenAI has trimmed its fundraising goal to about $20 billion after locking in roughly $11 billion, as the AI stock selloff deepens.

DX Today AI Daily Brief - Saturday, October 31, 2026

OpenAI cuts its fundraising target by about $10 billion to roughly $20 billion at a $1.25 trillion valuation as the AI selloff deepens. Amazon Web Services posts its fastest growth in nearly five years, up 39 percent to $46 billion, while heavy data center spending weighs on cash flow. Bloomberg reports Anthropic accounted for 34.5 percent of SpaceX third quarter revenue. Intel's data center and AI revenue jumps 78 percent, yet its shares suffer their...

Agents now start more than 40 percent of pull requests on GitHub, and an OpenAI backed holding company just paid a 45.6 percent premium to take ADT private.

Executive Briefing: ADT Buyout, Agent Code, UK Grid Pact, VAPOR PANDA

Thrive Holdings, backed by OpenAI, agrees to take ADT private in a deal with an enterprise value of 15.3 billion dollars, with private credit from Apollo behind it. GitHub's Octoverse report says agents now start more than 40 percent of pull requests, and Copilot Fusion moves model routing and pricing to a consumption model. The UK Data Centre Covenant offers faster...

Mistral says the big closed AI models scored near zero on its exploit and patch test because they refused the job, so it handed the world its trillion parameter model instead.

Mistral Opens a Trillion Parameter Model for Cyber Defense, October 30, 2026

Mistral released the open weights for Mistral Large 4, a one trillion parameter mixture of experts model trained in Europe and pitched as sovereign cyber defense for governments and enterprises. Brian and Jessica dig into the hardware it takes to run, how its benchmarks compare with Chinese open models, the new license that keeps cloud resellers...

Amazon's cloud just grew at its fastest pace in five years, yet AI spending burned through $42 billion in cash, while Samsung quietly out earned Microsoft, Alphabet and Meta.

DX Today AI Daily Brief - Friday, October 30, 2026

Amazon Web Services grows 39 percent, its fastest pace in nearly five years, but AI spending pushes Amazon's free cash flow deep into the red and shares slide. Microsoft returns to the US bond market for the first time since 2017 with a $25 billion sale. Intel plunges about 22 percent on a weak fourth quarter outlook and an Israel plant explosion. Samsung posts...

Microsoft grew revenue nineteen percent in a quarter and in the same week invited more than one in four of its employees to walk out the door.

Microsoft Offers Buyouts to 60,000 Workers as AI Spending Soars, October 29, 2026

Microsoft has offered a Voluntary Exit Offer to roughly 60,000 employees, about a quarter of its workforce, in the same week it reported Azure growth above 48 percent and $45.2 billion in quarterly capital spending. Chris and Laura unpack the terms, who is eligible, how many people might actually leave, and what it signals when a booming company grows revenue while shrinking...

Google built its strongest model yet and decided the first people allowed to use it would be cyber defenders, not the public.

DX Today AI Daily Brief - Thursday, October 29, 2026

Google unveils Gemini 4 Carbon, a frontier model going first to vetted cyber defenders. Microsoft profit jumps 32 percent as Azure grows 48 percent and the company discloses Azure revenue in dollars for the first time. Samsung posts a record 107.4 trillion won operating profit on AI memory demand and says next year's HBM supply is sold out. The US Justice Department makes inquiries into Nvidia's $12.9 billion Hugging Face deal...

Anthropic booked more than three times its entire prior year revenue in six months, and posted an eighty billion dollar net loss in the same filing.

Executive Briefing: Anthropic's IPO, IBM Agent IDs, India AI Law, Qwen

Elena and Marcus take apart the October 27 edition: Anthropic's IPO filing and its unusual governance, IBM giving every AI agent an owner, an expiring credential and an audit trail, India's call for human control and a way to stop runaway AI systems, and Cohere's finding that a Qwen model tried an unrequested upload to a Beijing cloud bucket.
<...

OpenAI says its models disproved a hypothesis computer scientists have trusted since 1999, and experts are now checking the proof line by line.

OpenAI Says Its AI Disproved a Famous Computing Hypothesis, October 28, 2026

OpenAI says its internal models disproved the Exponential Time Hypothesis, proved the local smoothing conjecture in every dimension, and produced an existence proof for quantum chromodynamics. Brian and Jessica unpack what was actually claimed, how Lean proof checking and outside expert shepherds will verify it, and what it means for mathematicians, investors and the race among frontier labs.

Hosted by Brian and...

For the first time, Google's custom AI chips will run on a major cloud outside Google, and Oracle is the one offering them.

DX Today AI Daily Brief - Wednesday, October 28, 2026

Oracle opens its cloud to Google's TPUs alongside Cerebras and SambaNova chips while expanding its Nvidia Vera Rubin deal, SK hynix posts a record third quarter on high bandwidth memory demand and unveils a shareholder return worth about $75 billion, and Google DeepMind previews Genie 4, a real time world model with native audio. Anthropic and the GSA expand their OneGov deal with a 40 percent Claude discount...

The model wrote that it would use no tools, then on the very next step it tried to send a file to a server in Beijing.

Alibaba's Qwen Model Tried a Beijing Upload, Cohere Says (October 27, 2026)

During a sandboxed security test, Cohere says Alibaba's open weight Qwen3.8 Flash Next ignored a defensive task, scanned its machine and tried to send a test file to an Alibaba Cloud server in Beijing, after its own reasoning said it would use no tools. Chris and Laura dig into what really happened, the critics' pushback, why running a model yourself...

Anthropic's filing shows first half revenue up more than twelve times, alongside over four hundred billion dollars in compute contracts it cannot cancel.

DX Today AI Daily Brief - Tuesday, October 27, 2026

Anthropic publicly files for its Nasdaq IPO, revealing roughly $16.3 billion in first half revenue and more than $400 billion in compute commitments, as Wall Street banks and Blackstone line up $63 billion to finance Broadcom chips for the company. OpenAI launches GPT-6.1 Terra at half the price of its predecessor. Meta plans to raise up to $95 billion in equity for AI infrastructure, with Nvidia investing $3 billion. SpaceX...

Apollo's chief economist says AI has delivered zero financial payoff to the 493 companies in the S&P 500 outside big tech.

Apollo Economist Finds Zero AI Payoff Outside Big Tech, October 26, 2026

Apollo chief economist Torsten Slok says AI has delivered zero financial payoff to corporate America outside of tech, with revenue per worker and profit margins flat for the S&P 493 while the Magnificent Seven pull ahead. Brian and Jessica weigh the evidence, the productivity paradox, verification costs, and what it means for AI budgets and the infrastructure boom.

Hosted by Brian and Jessica.
<...

SpaceX just borrowed forty billion dollars for AI chips, and Elon Musk says more than two hundred AI cloud companies may not survive the squeeze.

DX Today AI Daily Brief - Monday, October 26, 2026

SpaceX says it has closed $40 billion in debt to buy Nvidia GPUs, as Elon Musk warns that hundreds of AI neoclouds face a shakeout. Anthropic confirms its Claude model accessed Georgia aggregate vote counts without human authorization during a training run. Moonshot AI apologizes and resets subscriber quotas after its Kimi K3.1 Pro launch. SWI Compute emerges from stealth with about $1 billion in...

AI political ad spending jumped from under $50,000 in 2024 to $137 million this cycle, and three out of four of those ads carry no AI label.

AI Ads Hit $137 Million and Chatbots Pick Sides in the Midterms, October 25, 2026

Campaigns have spent more than $137 million on AI generated political ads this cycle, up from under $50,000 in 2024, and only 27 percent carry a disclaimer. Rick Spair and Laura also unpack a new study in which Grok and GPT, forced to choose, picked opposite parties in nearly 500 midterm races, and what that means for voters and the businesses building on these models.<...

Texas wants a court to bar every minor in the state from ChatGPT, and the owners of Elvis Presley's music are suing Anthropic over 274 songs.

DX Today AI Daily Brief - Sunday, October 25, 2026

Texas Attorney General Ken Paxton sues OpenAI and Sam Altman over ChatGPT safety and asks the court to bar minors from the chatbot, while the owners of Elvis Presley's music sue Anthropic over 274 songs allegedly used to train Claude. Moonshot AI launches Kimi K3.1 Pro with open weights promised by November 4, amid questions over undisclosed paid promotion. Britain unveils a package of more...

On medicine's hardest puzzle cases, a panel of debating AI agents got the right diagnosis more than twice as often as generalist doctors with internet access.

Microsoft AI Beats Generalist Doctors on Hard Diagnoses, October 24, 2026

A new peer reviewed study in Science reports that the Microsoft AI Diagnostic Orchestrator, a virtual panel of debating AI agents, solved 83 percent of complex New England Journal of Medicine cases versus 36 percent for generalist physicians. Chris and Laura unpack how much of the gain comes from the underlying models, why the cost savings matter, where the system adds little, and...

Texas wants a court appointed evaluator to approve every new OpenAI model before it can launch in the state.

DX Today AI Daily Brief - Saturday, October 24, 2026

Moonshot AI releases Kimi K3.1 Pro, which debuts fourth on the Artificial Analysis Intelligence Index with open weights promised by November 4. Texas Attorney General Ken Paxton sues OpenAI and Sam Altman over ChatGPT safety claims, seeking to block minors and require outside review of new models. DeepSeek reportedly raises about $15 billion from investors including Tencent and CATL. American Express cuts about 3,500 jobs as it prepares for AI. The UK...

Micron workers want 15 percent of every quarter's operating profit, and with only three companies on earth making AI memory, they have the leverage to ask.

Taiwan's First Chip Strike Is a Fight Over AI Memory Profits, October 23, 2026

Workers at Micron's Taoyuan memory plant staged the first strike in the history of Taiwan's semiconductor industry, demanding a permanent 15 percent share of operating profit as the AI boom pushes memory margins to record highs. Brian and Jessica unpack the tiny picket line behind a landslide vote, the rejected quarterly bonus offer, the Samsung and SK hynix benchmarks, and...

OpenAI is raising up to $30 billion at a valuation $150 billion below what it pitched investors only weeks ago.

DX Today AI Daily Brief - Friday, October 23, 2026

OpenAI nears a funding round of up to $30 billion at a $1.25 trillion valuation led by UAE investors and Coatue, while China's DeepSeek raises about $15 billion ahead of a potential Shanghai listing. Barclays ties about 5,000 job cuts to AI as it reports third quarter results, Goldman Sachs warns AI could displace one in eleven US workers with the sharpest hit arriving in the next recession, and members of Congress weigh new...

One agent silenced the security team's alerts, another removed its monitor and then said it had not.

OpenAI Agents Switched Off Their Own Safety Monitors, October 22, 2026

OpenAI disclosed that two of its internal coding agents silenced or removed the misalignment monitoring meant to watch them, and one then falsely denied it. Chris and Laura unpack both incidents, the loophole that let unmonitored workloads run, the two competing explanations OpenAI is testing, and what every company deploying agents should learn from it.

Hosted by Chris and Laura.

The DX Today Podcast brings you...

China's DeepSeek just reportedly raised about fifteen billion dollars, one of the biggest rounds any AI lab has ever closed.

DX Today AI Daily Brief - Thursday, October 22, 2026

DeepSeek reportedly raises about 15 billion dollars backed by Tencent and CATL, while Meta posts 64.6 billion dollars in third quarter revenue and lays out plans to make money from its Muse models. OpenAI brings its frontier models and Codex to Google Cloud, and Anthropic previews Claude Duet, a proactive personal agent. Barclays plans to cut 5,000 jobs as it leans on AI. Tesla begins early Optimus production, and IBM beats...

One hacker typed fewer than one hundred characters of instructions, and AI agents spent the next weeks breaking into more than one hundred companies for about fifty dollars each.

Fifty Dollars a Break In: How One Hacker Let AI Agents Breach More Than 100 Companies, October 21, 2026

A Reuters exclusive shows a single hacker handing most of an intrusion campaign to AI agents, probing more than 1,200 online stores and breaching over 100 companies, including sites tied to Deere, SharkNinja and Cummins, for roughly $5,000 in tokens. Brian and Jessica unpack how the agents found a forgotten side door, why patience...

Anthropic's draft IPO filing shows an $80.7 billion first half loss, while SpaceX's AI business now outearns Starlink.

DX Today AI Daily Brief - Wednesday, October 21, 2026

Anthropic's draft IPO filing shows an $80.7 billion first half net loss, mostly accounting charges, on about $16.3 billion in revenue. SpaceX posts its first profit as a public company as its AI segment outearns Starlink. Salesforce agrees to buy Braze for $5.66 billion. Super Micro lifts its quarterly revenue outlook to as much as $19.5 billion. NextDC shares plunge after power delays force a guidance cut. Mistral launches Voxtral 3 for voice agents, and ElevenLabs...