AI Stocks Lead Market Meltdown | StrictlyVC Summary

1. Key Themes

Theme: The AI trade is cracking, and the cracks are spreading from public markets to the IPO pipeline and the physical buildout.

AI-led selloff hits the public markets

The week's top story is a sharp repricing of AI megacaps.

"Stock futures tanked Thursday evening after a brutal AI-led selloff sent the Nasdaq down more than 10% from its recent high and into correction territory. The Magnificent Seven lost more than $1 trillion in market value, with Nvidia alone shedding more than $440 billion."

The correction is already reaching the AI IPO window

Anthropic's planned listing is the first big casualty of sentiment. Investors expect to haggle on price or timing.

"Major investors planning to buy into Anthropic's IPO say its expected $1.5 trillion valuation may need to fall sharply after this week's AI stock rout, with one suggesting a figure closer to $1 trillion and another raising the possibility of delaying the offering."

Political backlash against data centers is now a hard constraint

Voters, not just regulators, are pushing back on infrastructure that AI economics depend on, and OpenAI is already adjusting.

"OpenAI is promising a more transparent approach to future data centers it owns and operates, including publishing construction plans, soliciting community input, protecting utility ratepayers, and avoiding nondisclosure agreements with public officials after voters overwhelmingly backed restrictions on data center development in this week's midterms."

The Journal's election-night anecdote underscores the disconnect between AI leadership and public sentiment.

"Over wine and hors d'oeuvres, executives in tuxedos watched results roll in showing voters across the country turning against the data-center buildout their industry depends on."

Theme: Capital is still flowing to physical-world autonomy, robotics, and defense at very large valuations.

Autonomous logistics and drones command mega-rounds

"Zipline... raised a $1.5 billion round at a $21.5 billion post-money valuation. The deal was co-led by Paradigm, WCM Investment Management, and Layer Global."

"Humble Robotics, a one-year-old San Francisco startup that develops autonomous electric trucks for commercial freight routes under 250 miles, raised a $51 million Series A round led by Obvious Ventures."

Defense tech is consolidating with huge checks

"Shield AI... agreed to acquire Echodyne... in a deal valued at up to $5.3 billion. Shield AI is raising $2.5 billion to help finance the purchase, including $1 billion in new funding at a $28.8 billion valuation."

Humanoid robotics reaches public markets via SPAC, with thin fundamentals

"Agility Robotics is set to become the first U.S. humanoid-robotics company to trade publicly, following a SPAC merger that values the Oregon company at about $2.5 billion... Agility generated just $1.8 million in revenue last year while posting a $138.1 million net loss."

Theme: Cybersecurity and AI-security remain a hot, well-funded category, with early-stage now out-raising late-stage.

Early-stage cyber is out-raising later-stage for the first time in over a decade

"He's betting on himself at a moment when early-stage cybersecurity startups are raising more money than later-stage rivals for the first time in more than a decade."

Large rounds and fast exits validate the category

"Cyera... is reportedly planning to raise around $800 million in a round led by previous investor Cyberstarts at a $20 billion valuation."

"SGNL, the identity security startup CrowdStrike agreed in January to buy for $740 million, 11 months after Brightmind led its Series A."

Agent security and payments are emerging sub-niches

"Snaplii... lets AI agents spend from capped, pre-funded wallets without accessing users' primary card details, raised a $15 million round."

Theme: Vertical AI agents are automating back-office work across insurance, construction, accounting, and finance.

Many small, specialized AI-agent rounds

"Kalepa... uses AI to automate and coordinate underwriting, pricing, and portfolio-management work for insurers, raised a $75 million Series B at a $470 million post-money valuation."

"Numica, a one-year-old New York startup that acquires small U.S. accounting practices and uses AI to automate bookkeeping, billing, tax filing, and payroll, raised a $21.4 million round... It also secured $21 million in combined equity and debt financing."

"Shackleton AI... uses AI agents to automate administrative work for construction firms and developers across project planning, management, finance, and communications, raised a $5 million seed round led by Battery Ventures."

"Finpilot... uses AI agents to automate investment research, reporting, portfolio monitoring, and due diligence for institutional investors, raised a $10 million Series A round."

2. Contrarian Perspectives

"Sell on the way up": a VC who says not every company is an IPO candidate

Most VCs publicly claim every portfolio company is headed to an IPO, while privately modeling exits. Ward says the opposite openly, and has the SGNL outcome to back it up.

"Every time venture capitalists write a check, they're secretly running numbers in their heads about different exit scenarios. It's their job as fiduciaries. But nearly all will tell you (publicly, anyway) that every portfolio company is headed for an IPO. Stephen Ward... says otherwise outright. 'You have to sell on the way up,' he says."

Evidence: Brightmind led SGNL's Series A and saw it acquired for $740 million within 11 months of that round.

Burry: Anthropic is "betting its entire existence" on AI growth

Against the narrative of AI labs as unassailable, Burry flags take-or-pay compute commitments as a structural fragility that could force payment for unneeded capacity.

"Michael Burry says Anthropic is 'betting its entire existence' on continued AI growth, pointing to the company's $895.4 billion in future compute commitments and warning that take-or-pay contracts could leave it paying for capacity it doesn't need. The 'Big Short' investor says he has no plans to buy Anthropic shares when it goes public."

Safety watchdog says Anthropic's own risk disclosure was too reassuring

A third-party evaluator is challenging the lab's self-reporting, even while rating immediate risk as low.

"In a report published today, METR says Anthropic may already have crossed a key safety threshold, with AI nearly doubling the pace of its own AI research. The nonprofit also concluded that Anthropic's public risk report was too reassuring and that some of METR's analysis was redacted, though it still judged the immediate catastrophic risk to be low."

3. Companies Identified

Anthropic

Description: Frontier AI lab preparing for an IPO.
Why mentioned: Its IPO valuation is under pressure, and Burry and METR both raised concerns.
Quotes: "Major investors planning to buy into Anthropic's IPO say its expected $1.5 trillion valuation may need to fall sharply after this week's AI stock rout." / "Anthropic is 'betting its entire existence' on continued AI growth."

Nvidia

Description: AI chip leader.
Why mentioned: Biggest single loser in the selloff.
Quotes: "Nvidia alone shedding more than $440 billion."

OpenAI

Description: AI lab and data center operator.
Why mentioned: Responding to the data center backlash, and hired Cat Wu.
Quotes: "OpenAI is promising a more transparent approach to future data centers it owns and operates."

Brightmind Partners

Description: Two-year-old New York cybersecurity VC fund.
Why mentioned: Case study in disciplined exits, with a $250M debut fund and a second fund being raised.
Quotes: "Ward... says otherwise outright. 'You have to sell on the way up.'" / "SGNL... 11 months after Brightmind led its Series A."

CrowdStrike / SGNL

Description: Public cybersecurity leader and the identity security startup it acquired.
Why mentioned: $740M acquisition proof point.
Quotes: "SGNL, the identity security startup CrowdStrike agreed in January to buy for $740 million."

Cyera

Description: Data security startup.
Why mentioned: Reported $800M raise at a $20B valuation.
Quotes: "Cyera... is reportedly planning to raise around $800 million... at a $20 billion valuation."

Zipline

Description: Autonomous drone delivery company.
Why mentioned: Largest round in the issue.
Quotes: "raised a $1.5 billion round at a $21.5 billion post-money valuation."

Shield AI / Echodyne

Description: Defense AI/drone maker and radar company.
Why mentioned: Up-to-$5.3B acquisition in defense tech.
Quotes: "agreed to acquire Echodyne... in a deal valued at up to $5.3 billion."

Wonder

Description: Food-tech company founded by Marc Lore.
Why mentioned: Confidential IPO filing.
Quotes: "has confidentially filed for an IPO after raising more than $650 million in July at a roughly $9 billion valuation."

Agility Robotics

Description: Humanoid robot maker.
Why mentioned: First U.S. humanoid company to go public, via SPAC, despite minimal revenue.
Quotes: "Agility generated just $1.8 million in revenue last year while posting a $138.1 million net loss."

Dropbox / Tigris Data

Description: Cloud storage company acquiring a storage startup.
Why mentioned: Largest Dropbox acquisition, aimed at AI agent infrastructure.
Quotes: "Dropbox is buying cloud storage startup Tigris Data for $350 million."

Kalepa

Description: AI underwriting automation for insurers.
Why mentioned: $75M Series B at $470M post-money.
Quotes: "raised a $75 million Series B at a $470 million post-money valuation led by Updata Partners."

Harbinger

Description: Electric and hybrid vehicle platform maker.
Why mentioned: $240M Series D.
Quotes: "raised a $240 million Series D co-led by Tiger Global, Capricorn's Technology Impact Fund, and Ridgeline."

Ocient

Description: Petabyte-scale data analytics for AI agents and analysts.
Why mentioned: $100M Series C.
Quotes: "raised a $100 million Series C round co-led by BOKA Capital and OCA Ventures."

Alethio Therapeutics / T-Therapeutics

Description: UK antibody and T-cell therapy biotechs.
Why mentioned: Sizable biotech rounds.
Quotes: "raised a $90 million Series A round co-led by Sofinnova Partners and RA Capital." / "raised a $133 million Series B round co-led by Advent Life Sciences and British Business Bank."

Humble Robotics

Description: Autonomous electric trucks for short-haul freight.
Why mentioned: $51M Series A one year after founding.
Quotes: "raised a $51 million Series A round led by Obvious Ventures."

Liquid AI / Mercedes-Benz

Description: MIT spinout building on-device AI, and its automaker customer.
Why mentioned: On-device conversational AI deployed in cars.
Quotes: "conversational AI from Liquid AI... that runs directly in the car rather than the cloud."

Block

Description: Fintech parent of Cash App and Square.
Why mentioned: Another round of layoffs after earlier cuts.
Quotes: "Block has begun another round of layoffs just eight months after Jack Dorsey cut more than 4,000 jobs."

Numica

Description: AI-enabled accounting roll-up.
Why mentioned: Combines acquisitions with automation.
Quotes: "acquires small U.S. accounting practices and uses AI to automate bookkeeping, billing, tax filing, and payroll."

Firmly

Description: Connects retailer systems to AI shopping assistants.
Why mentioned: Agentic commerce infrastructure, with payments backers.
Quotes: "connects retailers' catalogs, inventory, payments, and checkout systems with AI shopping assistants."

Snaplii

Description: Fintech for AI-agent spending.
Why mentioned: Safe agent payments.
Quotes: "lets AI agents spend from capped, pre-funded wallets without accessing users' primary card details."

Lab-grown meat sector

Description: Cultivated meat startups.
Why mentioned: Cautionary tale on hype versus commercialization.
Quotes: "Lab-grown meat drew about $3 billion in investment after its 2013 debut, but today it is sold at just one store worldwide."

4. People Identified

Stephen Ward

Description: Co-founder and now sole GP of Brightmind Partners; ex-Coast Guard, nine years in the Secret Service, former CISO at TIAA and Home Depot.
Why mentioned: Operator-turned-investor raising a second $250M cyber fund solo.
Quotes: "Ward is raising a second fund, targeting roughly the same size as the first one." / "He's betting on himself... leaving Ward as the firm's sole general partner."

Dario Amodei

Description: Anthropic CEO.
Why mentioned: NYT profile on the worldview behind the company.
Quotes: "rationalism, effective altruism, early fears about AI wiping out humanity, and a conviction that the technology must be built 'right' shaped both his worldview and the company he co-founded."

Jensen Huang

Description: Nvidia CEO.
Why mentioned: Hosted AI leaders at the opera on election night as voters turned on data centers.
Quotes: "hosted what The Wall Street Journal described as 'the biggest names in America's artificial-intelligence boom' at the San Francisco Opera House."

Michael Burry

Description: "Big Short" investor.
Why mentioned: Bearish on Anthropic's compute commitments.
Quotes: "take-or-pay contracts could leave it paying for capacity it doesn't need."

Cat Wu

Description: Former Anthropic product leader.
Why mentioned: Talent move to OpenAI.
Quotes: "former Anthropic wunderkind Cat Wu has joined OpenAI to lead product management for its Core products."

Jason Spinell

Description: Early-stage investor; helped launch the Slack Fund in 2016.
Why mentioned: Joined Collaborative Fund as partner.
Quotes: "led investments in nearly 100 early-stage companies including Lattice, Loom, Glean, Hightouch, 1Password, and Fundamental."

Marc Lore

Description: Founder of Wonder.
Why mentioned: Taking Wonder public.
Quotes: "Wonder, the eight-year-old food-tech company founded by Marc Lore, has confidentially filed for an IPO."

Jack Dorsey

Description: Block CEO.
Why mentioned: Repeating layoffs he previously criticized.
Quotes: "repeated rounds of cuts were 'destructive to morale.'"

5. Operating Insights

Plan the exit from day one and sell into strength

Ward's rule is a counterweight to IPO-or-bust thinking, and it paid off quickly on SGNL.

"'You have to sell on the way up,' he says."

Pre-empt community and regulatory backlash on physical infrastructure

OpenAI's playbook for data centers is a template for anyone whose growth depends on local permission: transparency, community input, ratepayer protection, and no NDAs with officials.

"publishing construction plans, soliciting community input, protecting utility ratepayers, and avoiding nondisclosure agreements with public officials."

Operator credibility is a fundraising and deal-sourcing edge

Ward's background as a CISO and law enforcement officer differentiates his fund in a crowded category.

"A third reason is his resume, which includes a stint with the Coast Guard, nine years in the Secret Service, and CISO jobs at TIAA and Home Depot before he ever thought about investing."

6. Overlooked Insights

California's anti-wealth-tax measure may foreclose future versions

While the 5% billionaire tax is losing, a second measure could permanently limit similar taxes, which matters for founder and investor tax planning.

"Proposition 42 — a separate measure that would bar new taxes on individually owned assets and retroactive taxes, potentially blocking future versions of a billionaire wealth tax — seems on track to pass."

AI may already be nearly doubling the pace of AI research

METR's finding implies recursive acceleration inside frontier labs, which bears on the timing assumptions behind compute commitments and valuations.

"AI nearly doubling the pace of its own AI research."