This website uses cookies

Read our Privacy policy and Terms of use for more information.

In partnership with

Anthropic makes its public-market debut today as OpenAI releases a model it says could speed up AI research, and a government report puts four chatbots’ harmful capabilities to the test.

Introduction: Thursday’s Market Open Has a Second Story

A reader refreshes a market page before the workday begins.

One number will tell them how public investors value Anthropic. Another question is harder to price: what happens when AI systems become more capable, more widely used, and harder to govern?

That is the tension in today’s headlines. Anthropic’s Nasdaq debut is a financial milestone. OpenAI’s new model has reopened the debate over whether AI can accelerate its own development. And a Government Accountability Office report says investigators got four leading chatbots to produce workable attack plans.

These stories belong together. Not because they prove one sweeping conclusion, but because they show AI’s stakes widening all at once.

200 Ways To Make Money With AI

Ready to transform artificial intelligence from a buzzword into your personal revenue generator?

HubSpot’s groundbreaking guide "200+ AI-Powered Income Ideas" is your gateway to financial innovation in the digital age.

Inside you'll discover:

  • A curated collection of 200+ profitable opportunities spanning content creation, e-commerce, gaming, and emerging digital markets, each vetted for real-world potential

  • Step-by-step implementation guides designed for beginners, making AI accessible regardless of your technical background

  • Cutting-edge strategies aligned with current market trends, ensuring your ventures stay ahead of the curve

Download your guide today and unlock a future where artificial intelligence powers your success. Your next income stream is waiting.

Anthropic’s IPO Puts a Huge Number on AI’s Ambition

Last night, Anthropic priced its IPO at $83 a share.

That is above the $70-to-$78 range it set last week. At $83, the company is worth about $1.5 trillion on a fully diluted basis. Shares start trading today on Nasdaq under the ticker ANT.

The climb was steep. Anthropic was valued at $965 billion in a funding round in May. Its public filing landed last month. By Wednesday, demand was strong enough that reports had the price heading past $80, and it did.

The deal has an unusual feature. Seven of Anthropic’s chip and cloud suppliers: Google, Nvidia, AMD, Broadcom, Micron, Microsoft, and TSMC agreed to buy $26 billion of non-voting shares in a private placement alongside the IPO. Google and Nvidia are each in for roughly $10 billion. The companies that sell Anthropic its compute are now also their shareholders.

A public listing turns private expectations into a price that moves in real time. It does not settle what the company is worth over the long run.

A headline valuation is a starting point for market debate, not a verdict on future performance.

OpenAI’s New Model Raises a Different Kind of Valuation Question

While Anthropic heads to market, OpenAI shipped GPT-6.2 Astra and Sol on Tuesday.

Astra, the flagship, costs 20% less than GPT-6 Astra through the API. But the sharpest debate is not about price. It is about OpenAI’s own assessment of the model.

OpenAI says Astra is the first model it has released to meet the “High” threshold for AI self-improvement under its Preparedness Framework. The company’s analogy: it is like giving every OpenAI researcher an extra, highly capable research engineer.

That is a consequential claim. There is no evidence that Astra can build its own successor or run a research program alone. It is a risk rating, a signal about what OpenAI believes it must test, monitor, and explain.

The reception is split

Early testers are not all impressed. Developer Theo Browne wrote: “This model is clearly super smart. It is insufferable to work with on real projects.” He said he would not choose it over Claude Opus 5.5 for his own work, and that Astra spent seven days on a compiler rewrite and made it 56% slower.

One developer’s experience is not a benchmark. But it captures a real gap. “More capable” and “better for my workflow” are not the same thing.

The GAO Report Gives the Safety Debate a Concrete Test

On Wednesday, the Government Accountability Office published the results of a covert test.

Investigators posed as would-be attackers. They asked ChatGPT, Claude, Gemini, and Grok for help. According to GAO, each chatbot produced a plan it judged viable for a bombing and for an attack on the power grid, in under 12 hours. Senator Maggie Hassan requested the study.

The findings are serious and also narrow. GAO tested each platform on a single scenario and says the sample is not statistically significant. It will not say which chatbot produced which plan.

The companies pushed back on timing. OpenAI said the test “took place months ago on an older version of ChatGPT.” Google said it used “earlier models.” Anthropic did not dispute that “a determined tester can get harmful content” from its system.

The practical takeaway is less dramatic but more useful. AI safety is not only about what a model does in a designed demonstration. It is about safeguards, access, monitoring, and how fast institutions respond when harmful use is attempted.

Nvidia’s China Exit Adds Another Pressure Point

Nvidia reported earnings Tuesday, its first under interim CEO Colette Kress.

Revenue hit $115.8 billion, well above the $109.1 billion analysts expected. Data-center revenue alone was $107.6 billion. Guidance for next quarter: $130 billion.

The bigger news was China. Kress said Nvidia is no longer shipping AI chips there, despite holding U.S. export licenses. China sales are excluded from the outlook entirely.

And the leadership question got an answer. Jensen Huang said he would be “honored” to return as CEO if asked. On the earnings call, Nvidia’s board chair ruled it out.

The Trade-Offs Behind Today’s Headlines

Potential upsides

•    A public listing gives investors a clearer way to assess Anthropic’s value.

•    Lower model prices make useful AI tools more accessible.

•    Self-improvement thresholds clarify which capabilities developers are watching.

•    Government testing exposes vulnerabilities that might otherwise stay hidden.

Real concerns

•    A $1.5 trillion valuation sets expectations before the long-term economics are proven.

•    Cheaper, more capable tools also lower the barrier for harmful use.

•    A safety rating is not a guarantee that every risky behavior has been prevented.

•    Technology can outpace the rules and response plans of companies and governments.

FAQs Readers Are Asking

When does Anthropic start trading?

Today, Thursday, November 19, on Nasdaq under the ticker ANT.

What was the final IPO price?

$83 per share, above the $70-to-$78 range. That values Anthropic at about $1.5 trillion fully diluted.

What does OpenAI’s “High” self-improvement threshold mean?

It is a risk category in OpenAI’s Preparedness Framework, tied to a model’s potential to materially speed up AI research. It is an assessment, not proof the model can build a more advanced system on its own.

Did all four chatbots produce attack plans?

GAO says yes, one scenario each, all four judged viable. GAO also says the sample is too small to be statistically significant, and the companies say older model versions were tested.

Should the GAO findings be treated as proof of real-world attacks?

No. They show a serious capability and safeguards concern. They do not show how often such outputs occur or whether any plan was carried out.

Key Takeaways

•    Anthropic priced at $83, above range, and trades today as ANT at roughly $1.5 trillion.

•    GPT-6.2 Astra pairs a 20% price cut with OpenAI’s first “High” self-improvement rating.

•    The GAO report makes harmful chatbot outputs a concrete governance issue with real limits on what it proves.

•    Nvidia beat on earnings, exited China, and closed the door on Huang’s return.

Conclusion: One Screen, Several Kinds of Risk

Back to the reader refreshing the market page.

The IPO price is the number they can see. The harder questions sit just past it. Can AI companies meet investor expectations? Can safety practices keep pace with new capabilities? Can governments test systems before harmful use become a real-world problem?

Today’s headline is a stock-market debut. The fuller story is how fast AI’s financial, technical, and public-safety stakes are converging.

As you follow the opening trade, pick one company announcement or safety claim and check it against the original filing or report. Then share what you find with someone tracking AI’s impact.

Thanks for being a valued subscriber.

Pete Nyandeh,

Founder: AI Daily Brief, aidailybrief.io

Sources

•     TechCrunch, “Anthropic reportedly prices IPO at $83 a share, above projected range,” November 18, 2026

•     Yahoo Finance, “Anthropic Sets IPO Price Range at $70 to $78, Valuing Company at Up to $1.4 Trillion”

•     The Next Web, “OpenAI’s GPT-6.2 Astra cuts prices 20%, but early testers push back,” November 18, 2026

•     Fortune, “Washington just caught 4 AI chatbots planning terror attacks,” November 18, 2026 (GAO report)

•     Forbes, “Nvidia Reports Strong $115.8 Billion Quarterly Revenue In First Earnings Post-Jensen Huang,” November 17, 2026

•     Bloomberg, “Interim Nvidia CEO Says It’s No Longer Sending AI Chips to China,” November 18, 2026

•     Decrypt, “Jensen Huang Says He’d Be ‘Honored’ to Return as Nvidia CEO, Despite Board’s Resistance”

Disclosure & Disclaimer

This newsletter is for informational purposes only and is not investment, legal, or security advice. Figures reflect reporting available at publication and may be updated.

Reply

Avatar

or to participate