AI Spending and Copyright Risk Converge on Meta and Perplexity

Two Fronts, One Reckoning

The artificial intelligence boom has mostly been told as a story of momentum: more money, more models, more products. Two recent developments point to a less flattering picture. One concerns how much companies are spending on AI. The other concerns whether they had the right to the data that powers it.

At Meta, Mark Zuckerberg is reportedly pressing ahead with enormous AI investments despite having little to show for them so far.1 Meanwhile, in a federal courtroom in New York, the AI search startup Perplexity AI and its partner, the web-scraping service SerpAPI, failed to get key Digital Millennium Copyright Act claims thrown out of a lawsuit brought by Reddit.2

The two stories involve different companies and different problems. Read together, they suggest that the AI industry's two biggest inputs, capital and content, are both becoming more expensive and harder to justify.

Meta's Costly Conviction

The criticism of Meta is blunt. The company's spending on AI has been massive, yet it has produced almost nothing tangible to point to, and Zuckerberg is doubling down anyway.1 That framing casts the strategy as a high-stakes bet that has not yet delivered.

There is a familiar pattern here. Zuckerberg has a history of making sweeping, company-defining pivots and sticking with them through public skepticism. Supporters see that persistence as long-term vision. Critics see a CEO who can keep spending because his control over the company shields him from investor pushback. The available reporting does not settle which view is right. It does make clear that the gap between spending and visible results has become the main lens through which Meta's AI push is now judged.1

In my view, the harder question for Meta is not whether it can afford to keep spending. It is whether a clear product payoff arrives before patience runs out.

Perplexity's Legal Setback

The Perplexity case shows a different kind of cost. U.S. District Judge Paul A. Engelmayer declined to dismiss DMCA claims against Perplexity and SerpAPI in Reddit's copyright lawsuit, a ruling described as a substantial blow to the defendants.2 The decision does not resolve the case. It does mean Reddit's claims will move forward instead of being cut off early.

The broader context is a growing collision between AI developers and established copyright principles, with outcomes that could reshape how information is created, licensed, and consumed.2 Perplexity's business depends on gathering and summarizing web content. A legal theory that makes scraping, or the way scraped data is obtained, actionable under the DMCA goes straight to the heart of that model.

The involvement of SerpAPI matters too. By naming a scraping intermediary alongside the AI company, the case raises the possibility that liability can extend across the data supply chain, not just to the firm whose product reaches users.2 That should concern any AI company that relies on third-party tools to collect web data.

Where the Stories Converge

These stories come from different corners of the industry, and the coverage takes different tones. The Meta critique is openly skeptical and focuses on returns.1 The Perplexity coverage is more legal and procedural, concerned with what one ruling means for a wider body of law.2 Neither directly addresses the other.

Still, the overlap is worth noting. For several years, the AI sector has worked on two assumptions: that heavy capital spending would eventually pay off, and that the open web was largely free raw material. Meta's situation challenges the first. Perplexity's challenges the second.

If training and serving AI models requires huge upfront investment, and the data feeding those models brings growing legal exposure, the economics get tighter from both directions. Large incumbents like Meta can absorb uncertain returns for longer. Smaller players like Perplexity have less room for courtroom losses that might force licensing deals or limit how they gather information.

The Takeaway

It would be premature to call either development a turning point. Meta may yet turn its spending into products that justify it, and Perplexity may still win on the merits. What both suggest is that the AI industry is moving out of a phase where ambition alone was enough. Investors want results, and courts are starting to scrutinize where the data came from.

The companies best placed for the next phase will likely be those that can show both a credible path to revenue and a defensible claim to their data. For now, Meta faces doubts about the first, and Perplexity faces a serious test of the second.