Arena

Tesla is hiring a Machine Learning Engineer for Realtime Games, Digital Optimus, in Palo Alto. The job: build agents that play AAA games, browse the web and use pro apps at superhuman levels. Why games? Digital Optimus, the Tesla and xAI project Musk announced in March, reads screen video and acts with a keyboard and mouse. AAA games are a safe, endless, real-time proving ground for exactly that, the same approach DeepMind's SIMA uses. Tesla's posting says it will benchmark how skills transfer across games, general computer use, FSD and Optimus. tesla.com/careers/search/job…

Neoclouds are CRASHING 10% after Texas turned blue and its new governor threatened to pause data center construction. $NBIS has ZERO MW in Texas. Not online. Not in the pipeline. NOTHING. Its U.S. pipeline is elsewhere: • Independence, Missouri: 1.2 GW • Pennsylvania: up to 1.2 GW • Birmingham, Alabama: 300 MW • Vineland, New Jersey: 300 MW And it has something no other neocloud has: a real European footprint. Finland, France, the UK, Iceland, Israel, Spain. If the U.S. gets hostile to data centers, $NBIS is the hedge. In fact, a pause in Texas, home to 27% of planned U.S. data center capacity, is BULLISH for $NBIS. Less new supply = every MW already online is worth MORE. The market is selling the wrong stock.

#SouthKorea posts record $61.5 billion current-account surplus in September on #chip boom cnbctv18.com/economy/south-k…

Our security researchers built an autonomous cyber vulnerability discovery, exploitation, and patching pipeline. In a single run with an upcoming cyber model, it validated and generated patches for 948 local privilege escalation vulnerabilities in the Linux kernel at a total API cost of $1.2 million.

JUST IN: Microsoft cut its annual AI capital spending forecast by $5B after third-party data center delays. Satellite analysis suggests projects tied to IREN ($IREN) and Nebius ($NBIS), worth $27.1B combined, may be contributing. October 21 satellite images showed missing backup generators and cooling equipment at Horizon 4, one of four Texas facilities $IREN owns, raising concerns about its year-end deadline. Horizon 1 was previously delayed by equipment shortages but accepted in August. Research firm Aterio estimates roughly 140 MW at $NBIS New Jersey project could slip into 2027 following generator and fuel-cell permitting delays. Developer DataOne was fined $1M in September for operating generators without permits.

New York startups just raised $4.94 billion in October. Biggest funding month in the city's history. NYC even passed Silicon Valley in active AI office demand. We hit 2.79 million square feet compared to their 2.78 million. Founders want to build where capital and talent actually live.

$RDW: Redwire Q3 results are out w/ record backlog and record gross margins. Revenue $109.0M Gross margin 32.3% Backlog $619.7M Book-to-bill 1.73 EPS $(0.09), up $0.20 YoY Liquidity $538.9M Full-year revenue guidance of $450M to $500M is reaffirmed. The call is tomorrow at 9 AM ET.

New: Give all your agents one shared brain in Notion. Now, ChatGPT, Claude, Muse, and Bot can save what they learn to the same memory and use the same skills in Notion. Learn how to Notion-pill your agents:

Some news: I’m joining OpenAI to lead product management for Core Products! I have a lot of respect for OpenAI for showing the world that scaling works, for bringing AI to over a billion people, and for believing great products are core to the mission. I’m excited to work with @thsottiaux, an incredible PM team, and everyone across OpenAI. Together, we’ll build agents that have your back and your team’s, and keep you in control, so you get more time for the people and work you love. Lots to learn, and lots to build. Please keep your feedback coming!

$QCOM beat on revenue and EPS, and the stock fell more than 4% after hours anyway. Fiscal Q4 revenue was $10.32 billion against $10.11 billion expected, down 8% from a year ago. Adjusted EPS was $2.17 against $2.16. Smartphone chip revenue missed estimates, even after Qualcomm raised prices in September. The guide was mixed. For the December quarter, Qualcomm sees revenue of $10.4 to $11.2 billion, a midpoint above the $10.48 billion expected, but EPS of $2.15 to $2.35 against $2.27 expected. The reason is the same one that hit Apple and Intel this week: memory. AI servers are soaking up memory supply, prices are rising, and that raises costs for phone makers and squeezes demand. The Apple story is now baked in. CEO Cristiano Amon said the December guide includes no further Apple sales, after Qualcomm had already said Apple revenue would fall about 50% from the September to December quarter. The fiscal 2027 outlook assumes no Apple at all. Qualcomm’s answer is everything that isn’t a phone. Automotive hit a record near $2 billion. IoT grew 4% to $1.88 billion. Amon says non-handset revenue should grow more than 60% in fiscal 2027, with design wins up 70% excluding Apple. He reiterated a $5 billion data center target for fiscal 2027, with custom chip programs for two hyperscalers starting this quarter and a goal of $15 billion a year by fiscal 2029. The stock was already down about 9% over the week going into the report, and options had priced in a move of roughly 9%. The market is willing to pay for the non-handset story, but it wants to see it in reported numbers while the phone business, still the largest, shrinks.

📊 $SEZL EARNINGS RESULTS 🟢 $SEZL Sezzle Reports Q3 2026: EPS $1.38, Revenue $166.5M • 🟢 Revenue: $166.5M vs $131.93M expected (+26.2%), up 42.6% year over year • 🟢 EPS: $1.38 diluted vs $1.28 expected (+$0.10); adjusted EPS of $1.37 • 🟢 Guidance: full year 2026 outlook raised for the fourth time, with revenue growth now seen at about 42.5% and adjusted EPS of $5.45 to $5.50; preliminary 2027 adjusted EPS guide of $7.20 introduced • Standout: total transaction volume hit a record $1.4B, up 34.9% year over year, while the provision for credit losses eased to 2.7% of volume from 3.1% a year ago Can $SEZL keep credit losses contained through the holiday quarter, or does peak season volume come with a collections bill? 👇 NFA

Models by @GoogleDeepMind scale better on cost than their rivals. Gemini 4 Carbon is the prime example. As tasks get more demanding, every rival gets more expensive relative to Carbon. For a typical task, GPT-6 Astra by @OpenAI costs 3.4× as much, Claude Opus 5.5 by @AnthropicAI costs 3.9× as much, and GPT-6.1 Sol is the cheaper option at about half of Carbon’s cost (0.54×). Across the 5% most expensive sessions, Carbon’s position improves against all three. Opus edges up to 4.2×, Astra’s cost roughly doubles relative to Carbon, and Sol’s discount shrinks. This in turn can translate into meaningful savings across real-world tasks such as building an Android security app, one of many long-horizon tasks that our Agent Arena users delegate to their agents.

*CITRINI OVERHAULS THEMATIC MODEL BOOK, CALLS PEAK IN AI INFRASTRUCTURE CAPEX *CITRINI SAYS AI INFRASTRUCTURE ASYMMETRIC RISK HAS SHIFTED TO DOWNSIDE *CITRINI EXPECTS HIGHER COSTS FOR US DATACENTERS AFTER DEMOCRATIC MIDTERM WINS *CITRINI WARNS MACRO REGIME SHIFTING TOWARD LATE-CYCLE STAGFLATION TILTS

Anthropic's S-1 shows $17 billion in Q3 revenue, a $567 million operating loss, and compute commitments that run through 2045. So you can grow 1,100% in a year and still be on a payment plan until the 2040s. Ngl, that's the most AI thing ever.

Bro. Dude. Bro? Dutch Bros down 18% as comps slow badly. 4.4% is fine if your PE is 20. Less so at 60x. $BROS

$FIG Q3 results: → $391M in revenue, 43% YoY growth rate and $16M above high-end of prior guidance → 131% net dollar retention rate → 36% YoY growth in non-GAAP gross profit → $1.7B in cash, cash equivalents and marketable securities → Raising our outlook for FY 2026

Background subagents in Claude Code now use less of your usage limit. A running subagent updates its status line every 30 seconds, and each update used to re-read its whole conversation. Now Haiku 5.5 writes it from short notes instead, so each update costs about 99% less.

$UBER reported its Q3 earnings Wednesday morning: 💰 Revenue: $14.77 billion vs. $14.65 billion, expected 🪙 EPS: $0.88 vs. $0.87, expected 🚙 Gross bookings: $59.8 billion vs. $59.8 billion, expected 📈 Q4 gross bookings outlook: $63.25 billion to $65.25 ​billion vs. $64.4 billion, expected

NEW: Apple is cutting employee AI spending budgets even as it pushes workers to use the technology more. Engineers have seen daily limits cut from $300 to $200, with further reductions to $100 planned. Other teams could see limits fall to $25. bloomberg.com/news/articles/…

Cognition launched SWE-2 Flash, a coding model for Devin at 1/5 the price and ~3x the speed of SWE-2. Built on Thinking Machines' Inkling-Small (276B, 12B active). Weights are held back over cyber risk, shared only with vetted partners. spectrumweb.co/news/cognitio… #AI #LLM

$35M → $50M → $70M. Another month, another 40% MoM increase in annualized revenue across the Bittensor ecosystem. When we published the first SubConnect Bittensor Revenue Index, we estimated that revenue-generating subnets were collectively doing around $35M in annualized revenue. A month later, that number had grown to approximately $50M. Today, we're estimating $70M in annualized revenue. That's a 100% increase in just two months. Of course, annualized revenue is a snapshot, and the real challenge is sustaining that growth while improving margins and reducing dependence on emissions, but the direction is clear. Bittensor is growing from the inside out. And our expectation of $100M+ in annualized revenue by the end of 2026 is getting closer. We'll continue tracking the numbers every month through the SubConnect Bittensor Revenue Index.