The tech industry is hot for shopping bots, but they're years away

The underlying e-commerce infrastructure must change before AI can reliably make purchases on your behalf.

The tech and financial industries are enthusiastic about "agentic commerce," where AI agents make purchases on behalf of consumers. However, significant challenges remain, delaying widespread adoption. Major platforms like Amazon have already blocked bots from companies like Meta and Perplexity, citing terms of service violations. Payment companies, while seeing potential, acknowledge the need for fundamental changes in e-commerce infrastructure to support autonomous transactions. Experts estimate agentic commerce could reach $1 trillion in US consumer retail by 2030 and mediate over $15 trillion in B2B spending by 2028, but these figures are contingent on broad industry adoption and the resolution of technical and trust issues. Current e-commerce systems are designed for human interaction, not for the speed and determinism required by AI agents. Issues like identity verification, trust in AI decision-making, and the lack of transparency in purchasing processes are key hurdles.

Current e-commerce platforms are not built for AI agents, requiring a complete overhaul of user interfaces and backend systems to enable machine-speed transactions. This includes developing deterministic gates and new protocols to manage purchases, ensuring trust and transparency, and overcoming challenges related to identity verification and liability.

Despite projected market sizes, true agentic commerce is years away, with current capabilities limited to agent-assisted shopping. The industry needs to standardize how agents interact with merchants, handle negotiations, and manage settlements. Without these foundational changes, the vision of autonomous shopping bots remains a distant prospect, with humans still overseeing crucial purchase decisions.