Publishers are enthusiastic about marketplaces from AWS and Microsoft because they offer a path to usage-based revenue. This model is seen as more sustainable than the current one-off, flat-fee licensing deals with AI companies, potentially replicating the scalable monetization of digital advertising.
Workday's CEO change reflects a broader trend: the belief that founder-technologists are essential for navigating the AI transition. Similar to leaders who guided cloud migrations at Microsoft and Adobe, these founders are being brought back to ensure companies invest correctly and 'cross the chasm' in a post-AI world.
OpenAI's move into shopping features is hitting a major hurdle: sales tax complexity. This forces them to build a sophisticated backend for tax collection and remittance, similar to Amazon, rather than just being a thin AI layer. This will necessitate a hiring push for compliance and tax experts.
AWS's marketplace for publishers isn't just an AI play. It's a strategic move to compete with CDN providers like Cloudflare by innovating its CloudFront service. It also aims to build goodwill with publishers, which benefits Amazon's massive and growing advertising business.
The recent software stock drawdown is not about poor current performance; many companies are still beating earnings. Instead, the market is pricing in a massive "terminal value risk" from AI, valuing companies as if they will decline in perpetuity, creating a historic disconnect between current fundamentals and long-term valuation.
The next software market bounce will not lift all boats. An analyst predicts a "K-shaped recovery" where robust platforms (HubSpot, Microsoft) could see massive gains. In contrast, point-solution applications (Asana, DocuSign) are at high risk of being disrupted by AI features built into larger ecosystems.
AI agents like OpenClaw learn via "skills"—pre-written text instructions. While functional, this method is described as "janky" and a workaround. It exposes a core weakness of current AI: the lack of true continual learning. This limitation is so profound that new startups are rethinking AI architecture from scratch to solve it.