Sources: Reddit signed a deal early in 2024, worth ~$60M on an annualized basis, allowing a large unnamed AI company to train its models on Reddit's content
- Social media firm agrees $60 million deal with AI company — Reddit advised to seek at least $5 billion IPO valuation
Context & Ripple Effects
The reported agreement arrived as Reddit was targeting a public-market valuation in the mid-single-digit billions, making a recurring AI-data contract a potentially material part of the company’s pre-IPO revenue narrative. Reddit’s planned IPO valuation range provides the immediate financial backdrop.
Subsequent coverage identified Google as the counterparty and described access spanning model training and Reddit’s Data API. That turns the initial unnamed-buyer report into an early example of a platform commercializing its corpus for both AI development and product distribution. Google’s reported Reddit content agreement
First-order effects
- Reddit gains a reported annualized revenue stream of about $60 million from licensing its content for AI-model training, while the buyer gains authorized access to that corpus.
- The agreement gives Reddit an additional business metric for investors beyond its traditional platform operations as it approaches an IPO.
Second-order effects
- The reported deal establishes a concrete price reference for AI access to a large social platform’s content, increasing pressure on other content platforms to evaluate licensing rather than leave model-training access unmanaged.
- Because the later Google report includes Data API access and Search surfacing, the arrangement can tie content licensing to traffic-distribution relationships, not solely training-data sales. Google’s reported API and Search access illustrates that broader commercial bundle.
Third-order effects
- If such agreements become repeatable, user-generated-content platforms may treat archives and live discussion data as a distinct AI revenue line, reshaping negotiations over access, attribution, and product distribution.
- The pattern points toward a more permissioned market for high-value web content, where major AI developers secure direct platform relationships rather than relying only on open-web collection.
The trend: AI developers are moving toward direct licensing deals that turn platform content into a priced input for model training and AI-enabled distribution.