The AI agent arms race got its biggest challenger yet: Google launched Gemini Spark, a personal AI agent that reads emails, monitors credit card accounts, and can autonomously spend money — a direct shot at OpenClaw's territory a month after Peter Steinberger's $1.3M token bill made headlines. Meta is restructuring around AI at massive scale, laying off 8,000 employees while reassigning 7,000 to AI roles — a brutal but telling internal pivot. Anthropic hit a financial milestone: expected profitability for the first time since 2021, paired with a Claude Code update that makes it feel "less like a coding assistant and more like an autonomous software engineer." On the model frontier, Alibaba's Qwen3.7 Max jumped 4.8 points on the Artificial Analysis Intelligence Index — the closest a Chinese lab has gotten to the frontier, driven by better non-hallucination rates. And the OpenClaw ecosystem keeps expanding sideways: Fastlane launched "OpenClaw for Marketing" for one-click social media deployment, Clawnetes shipped a desktop wizard, and an independent audit confirmed that AI agents at major companies can "cheat, deceive, and work unsupervised" — safety concerns in production, not theory.
Google's new personal AI agent reads your emails, monitors credit card bills, and can autonomously spend money or send emails on your behalf. Power users have already automated inboxes, calendars, and texts; the Meta employee who nearly lost their entire email trove to an OpenClaw agent serves as a stark reminder that autonomous agents with spending power raise the stakes considerably.
Mark Zuckerberg is betting Meta's future on AI, shifting 7,000 employees to AI roles while simultaneously laying off ~8,000 (10% of the global workforce). The dual motion — massive AI reinvestment alongside deep cuts elsewhere — mirrors what Intuit is doing (cutting 17% to sharpen AI focus). Big Tech's AI pivot is no longer just CapEx; it's headcount.
An OpenAI model has disproved a central conjecture in discrete geometry — AI contributing original mathematical research, not just assisting human mathematicians. This is a different kind of milestone: AI as a primary discoverer in pure mathematics.
Anthropic is expected to reach profitability for the first time since its 2021 founding, while simultaneously shipping a major Claude Code update with remote session control, flicker-free terminal interfaces, and autonomous engineering capabilities. Marc Andreessen notes that world-class programmers started saying the new models are better than they are over the 2025 holiday break — "coding is first because AI companies are using it on themselves. Everything else follows."
Nvidia deployed cash across the AI infrastructure stack, funding businesses that could turn around and buy its technology — a self-reinforcing investment flywheel. Combined with Jensen Huang's recent "demand has gone parabolic" comment on agentic AI, the chipmaker is positioning itself as both supplier and kingmaker in the AI economy.
Selected posts from the edition, with links to the originals.
"Alibaba's new Qwen3.7 Max model scores 56.6 on the Artificial Analysis Intelligence Index, 4.8 points higher than Qwen3.6 Max Preview (51.8). While Alibaba still trails models from OpenAI, Anthropic and Google, Qwen3.7 Max is the closest they have been to the frontier."
View post on X ↗"🚨 One prompt inside Claude's Cowork feature just wiped someone's data from every major data broker. Zero forms filled by hand."
View post on X ↗"I never bought the voice AI hype. To me they all sounded the same, choppy, robotic, lost the second anyone went off script. Every demo I'd seen had the same pattern. The agent handled 'what are your hours' fine. Then a customer would ask something slightly off-script and..." (describes voice AI finally crossing the uncanny valley — thread earned 34 likes, 31 retweets)
View post on X ↗"Anthropic is expected to reach profitability for the first time since it was founded in 2021."
View post on X ↗"🚨 Anthropic just dropped a 20-minute walkthrough of everything new in Claude Code. And the updates make it feel less like a coding assistant and more like an autonomous software engineer. New capabilities include: remote session control, flicker-free terminal interfaces..."
View post on X ↗"Circle $CRCL co-founder Sean Neville's Catena Labs raises $30M to build regulated banking infrastructure for AI agents."
View post on X ↗(Fastlane): "fastlane just launched openclaw for marketing — a new ai agent powered feature. you chat with your favorite AI agent like Claude, give it a simple instruction like 'market my app' and it automatically handles the entire process: deploys and warms real social media accounts..."
View post on X ↗"Independent assessment finds AI agents at major companies can cheat, deceive, and work unsupervised — but lack the sophistication for a sustained takeover."
View post on X ↗"Marc Andreessen says the AI coding debate flipped over the 2025 holiday break: World-class programmers started saying the new models are better than they are. His point: coding is first because AI companies are using it on themselves. Everything else follows. @pmarca"
View post on X ↗(Nous Research): "Grab a Grok subscription → Proceed to auth w/ hermes model → Turn on x_search on hermes tools → Search real-time news on..." — Grok Build 0.1 now available for early access in Hermes Agent. xAI's Grok integrates with Nous Research's self-improving agent framework.
View post on X ↗