An AI company’s market value exceeded $1 trillion on its first day of trading.
In the same week, multiple U.S. banks quietly updated their terms: if you authorize AI agents to access your accounts and something goes wrong, you might have to bear the losses yourself.
AI is becoming the world’s most valuable business and shifting new risks onto ordinary people.
Anthropic launched on Nasdaq, raising a record-breaking amount of capital and closing its first day up 17%.
Meanwhile, U.S. banks including Citi and PNC have modified their online agreements so that losses incurred from compromised third-party AI agents authorized by customers may be the customer’s responsibility.
Today’s two crucial AI news stories highlight issues from the perspectives of AI company financing and personal financial safety.
(Note: SasaDaily uses various AI tools to produce content, including Anthropic’s Claude. The first story is a summary of reported figures and risks without expressing a positive or negative view.)
1. Anthropic’s IPO Gains 17% on Debut: What Does This Historic IPO Mean?
What were the key IPO figures?
According to Yahoo Finance, Forbes, and FinanceFeeds:
- Underwriting price: $83 per share, above the initial estimate of $70–78.
- Shares issued: 980 million Class A shares, including approximately 23.65 million shares sold by existing shareholders.
- Funds raised: $81.3 billion, surpassing SpaceX’s $75 billion raise in June this year, making it the largest IPO in history.
- Listing venue: Nasdaq, ticker symbol ANT.
How did the stock perform on day one?
Based on November 19 trading data organized by FinanceFeeds:
- Opening price: $111.50, about 34% above underwriting price.
- Highest price: $115.10.
- Lowest price: $92.72.
- Closing price: $97.17, 17.1% above underwriting price.
Using the number of shares listed on Nasdaq, the closing market capitalization was approximately $1.27 trillion.
The stock surged at open but dropped significantly by close, showing high volatility on the first day.
What did Anthropic say?
According to FinanceFeeds, CEO Dario Amodei stated, “A commitment to safety and public interest has been part of Anthropic’s DNA from the start.”
President Daniela Amodei said, “One major reason for going public is to involve more people in advancing AI responsibly.”
What are the risks?
According to Yahoo Finance’s analysis of the public filings:
- Third quarter revenue was about $17 billion, with an operating loss of $567 million; the previous quarter showed an operating profit of $974 million.
- Future spending commitments total roughly $895 billion, about $70 billion of which is due within a year; cash and short-term investments stood at roughly $132 billion at the end of September.
- Some analysts predict the company may return to operating losses in the next couple of quarters.
- Yahoo Finance analysis notes that since 2008, all top 12 U.S. IPOs had prices fall below their initial deal price within a year.
The public filing also highlighted that Anthropic has reached a safety policy risk threshold related to “AI accelerating its own research.”
What does this mean for ordinary people in Taiwan?
SasaDaily previously covered Anthropic’s offer to Australian retail investors on November 12. Their official listing now signals the start of large-scale fundraising from general investors for AI companies.
If you see headlines like “largest IPO ever” or “stock jumps on first day,” remember:
- First-day price changes don’t indicate long-term performance.
- The company is still running losses and has significant spending commitments, which are risks to consider.
- Be cautious of ads or private messages promising guaranteed access to popular new stocks—they may be scams.
This article summarizes publicly available information and does not constitute investment advice.
Sources:
- Yahoo Finance: Anthropic IPO live
- Forbes: Anthropic IPO Live
- FinanceFeeds: Anthropic (ANT) Closes Debut at $97.17
2. U.S. Banks Change Terms: If Your AI Agent Is Hacked, Are You on the Hook?
What changes did banks make?
Startup Fortune reported on November 18 that several major U.S. banks updated their online service terms:
- Citi: Revised its online user agreement. Citi states it wants to “emphasize that customers who choose to authorize third-party AI agents to access accounts are responsible for these actions.” The report explains that if login credentials are leaked or hacked through the AI tool, losses caused by criminals accessing the account will be borne by the customer.
- PNC: Changed terms on November 18.
- Citizens Bank: Sent a similar notice to customers within the same week.
- Wells Fargo: New online terms effective November 24 state that transactions made by authorized agents—including AI agents—are considered authorized “even if you did not intend or want those transactions.”
The report also notes that JPMorgan Chase and Bank of America took similar steps in October: Chase blocking external AI agents from directly accessing accounts, and Bank of America advising customers not to share online banking passwords with AI assistants.
What is a "third-party AI agent"?
While not formally defined in the report, examples include:
- Browser-based AI agents that log into websites and navigate pages for you.
- Shopping AI agents that help compare prices and place orders.
- Financial AI assistants that remember your login credentials and manage your accounts.
Is this legal?
There is currently no clear legal precedent.
According to the report, U.S. consumer protection rules limit customer liability for “unauthorized” transfers if reported promptly.
However, by defining transactions made by AI agents as “authorized,” banks may exclude them from this protection.
The interpretation has not yet been tested in courts or by U.S. regulators, and there has been no official regulatory stance on the matter.
Why are AI agents particularly risky?
The report cites two main concerns:
- AI agents might be tricked by hidden malicious commands on webpages to perform actions you did not request.
- If the AI stores your login credentials, a failure or breach with the tool exposes your accounts.
SasaDaily covered a similar incident on November 12 when Notion’s AI provider was hacked and conversation records leaked, emphasizing: never share your passwords with AI.
Should Taiwanese users worry?
The updated bank terms apply to U.S. banks and do not directly affect Taiwan.
Nonetheless, as AI begins operating financial accounts, responsibility clauses in terms of service highlighting who bears losses are becoming more common.
Wherever you are, the basic rule remains: never provide your online banking credentials to any AI tool.
Source:
3. Why Should These Two Stories Be Considered Together?
One story is about AI companies securing massive funding, while the other concerns everyday people facing new risks:
- Anthropic’s IPO: Investors are placing huge bets on AI’s future.
- Bank terms changes: AI agents enter daily life, but if things go wrong, users are often the ones held responsible first.
The benefits and risks of AI don’t necessarily fall on the same groups. When reading AI news, consider who ultimately absorbs the risk.
4. What Should Ordinary People Watch Out For Today?
If you’re interested in AI stocks:
- Read public information first to understand whether the company is profitable or facing heavy spending pressures.
- Don’t chase the hype just because of headlines like “largest ever” or “big gains on the first day.”
If you use AI tools with your finances:
- Never share your online banking credentials with any AI tool.
- Using AI to organize bills or compare prices is fine, but handle payments and transfers yourself.
- Before connecting any AI tool to your financial accounts, carefully review your bank’s terms and conditions.
Today’s Most Important Reminder
AI companies’ market values can surpass trillions in a day, but the safety of your wallet is still your responsibility.
Recommended Reading
- AI Quick Guide | Before Your AI Agent Buys Something, Set Spending Limits and Approval Steps
- AI Q&A | 2026/11/15: The Fed Calls AI the Top Financial Risk. Should I Sell All My AI Stocks?
- AI Q&A | 2026/09/16: Does Claude’s Use of QuickBooks/Drive Access Guarantee It Can’t See Confidential Company Data?