[Into the AI World] Anthropic Fails to Revive AI Spark... Shares Plunge on Second Day of Trading
Raised Record-Setting Amount, but Gave Back Gains
"Must Prove Strong Revenue Growth and Improved Profitability"
Apple Moves to Develop AI Agent to Rival Claude
Anthropic, which entered the Nasdaq after raising the largest amount of capital in history, gave up its gains on November 20 (local time), its second day of trading, closing down 8.50% at $88.91. This has heightened concerns about AI stocks. With Anthropic's share price failing to gain momentum despite hopes that the company would rekindle waning enthusiasm for AI, expectations for AI stocks also appear to be falling.
Experts said caution is growing as investors face a host of uncertainties, including AI infrastructure investment and profitability, financing conditions, and variables related to U.S. monetary policy. Concerns over the pace and cost of AI development have intensified, while risk aversion has added to uncertainty. The situation involving Nvidia CEO Jensen Huang, skepticism over overheated investment in AI infrastructure, and data center regulations following the Democratic Party's sweeping victory in the midterm elections have sent technology stocks broadly tumbling, and Anthropic appears to have been caught in the sell-off as well. U.S. President Donald Trump said at a press conference that day that American companies could face significant disadvantages if they build data centers overseas.
Experts said Anthropic is now facing a test of its ability to monetize its business. They explained that the company will have to demonstrate steep earnings growth for years to justify its astronomical valuation. Current market forecasts for Anthropic's revenue are $55 billion in 2026, $135 billion in 2027, and $240 billion in 2028. Kim Jaeim, a researcher at Hana Securities, said, "The three things to watch at Anthropic going forward are its revenue growth trajectory, the burden of computing investment and profitability, and whether it can maintain its competitive edge in enterprise AI." She added, "During the IPO roadshow, some investment banks projected revenue of up to $137 billion in 2027 and $280 billion in 2028. Whether Anthropic can meet these high growth expectations will be a key factor for its share price going forward."
Anthropic's final prospectus disclosed approximately $900.9 billion in long-term commitments related to computing and infrastructure. Its adjusted operating margin fell from 19% in the second quarter of this year to 15% in the third quarter. The decline was attributed mainly to increased costs from securing computing resources in the short term. How efficiently Anthropic can manage computing costs and improve profitability as revenue grows is expected to affect investor sentiment.
Key questions include whether Claude's competitive edge in coding can expand to enterprise work more broadly, and whether Anthropic can maintain its pricing competitiveness amid competition with OpenAI and Google. Competition remains intense, with Apple, for example, developing its own AI agent to rival Claude. This year, companies are expected to spend $77.3 billion on AI applications, of which coding and chatbots account for 44% and 32%, respectively, together making up 76% of total AI application spending. Alongside the growth potential of the AI market, it will be important to see how much market share and profitability Anthropic can secure over the long term.
Researcher Kim said, "Anthropic's share of spending on application programming interfaces (APIs) for enterprise AI models is 55%, far higher than OpenAI's 31%." She added, "Ultimately, whether the company can demonstrate both strong revenue growth and improved profitability over the medium to long term will be a key factor in justifying its high valuation." Jo Jaeun, a researcher at Daishin Securities, said, "The market is focused less on AI demand than on whether AI investment can continue beyond next year," adding, "The future direction of AI investment depends on capital expenditures by hyperscalers, or operators of large-scale data centers."
© The Asia Business Daily. All rights reserved. Unauthorized AI training and use prohibited.