Vikram Taneja, head of AT&T Ventures, argues that AI has fundamentally reshaped what makes an early-stage startup defensible. While the technology has democratized software development by lowering the bar for building products, it has simultaneously raised the stakes for what separates winners from the pack at seed stage.
Taneja's thesis centers on technical risk reframing. Five years ago, the ability to execute on an ambitious engineering roadmap was a core competitive moat for young companies. Today, any team with competent engineers and access to AI tools can ship functional software quickly. This flattens the playing field but also makes raw technical execution table stakes rather than a differentiator.
What matters now, according to Taneja's perspective, is defensibility through other vectors. Startups need to demonstrate unique datasets, network effects, regulatory positioning, or domain expertise that AI tools cannot easily replicate. A seed-stage company must show it has something structurally harder to copy than the software itself.
AT&T Ventures, which invests across enterprise software, IoT, and network infrastructure, has adjusted its diligence accordingly. The firm now scrutinizes founders' unfair advantages outside pure technical prowess. Can the team access proprietary data? Do they own relationships no competitor can easily duplicate? Is there a regulatory or compliance moat?
This shift has implications across the venture ecosystem. It suggests seed investors should lean harder on evaluating founder pedigree, market position, and operational leverage rather than betting on technical execution alone. For founders, it means shipping a working prototype is no longer enough to raise institutional capital. The story must include defensibility beyond product.
Taneja's comments reflect broader VC realism around AI commoditization. As large language models and generative AI tools mature, the venture bar for "technical founder" has shifted from rare engineering talent to founder-