AI has lowered the cost of building sophisticated software. A small team can now use frontier models to create in weeks what once demanded years of engineering. For venture firm Team8, that abundance creates a different scarcity: if competitors can access much of the same intelligence, the harder question is what remains valuable after the next model arrives.
Team8 has raised $365 million in new capital, split between a $265 million Team8 Capital III fund and more than $100 million reserved for follow-on investments in selected portfolio companies. The third fund will back Seed and Series A startups across cybersecurity, software infrastructure, fintech and digital health, bringing Team8’s assets under management to nearly $2 billion across eight funds since 2014.
The progression also shows how the platform has grown. Team8 Capital’s first fund was $160 million and its second reached $235 million. The newest fund had already produced an exit before final close after Palo Alto Networks’ reported acquisition of Koi Security in February.
When intelligence becomes rentable, context becomes the asset.
That idea shapes where Team8 wants to invest. Managing Partner Sarit Firon argues that access to an underlying model is unlikely to provide a lasting advantage when competitors can use the same technology. Team8 instead expects durable companies to become embedded in critical workflows, customer trust, proprietary context, enterprise operations, and infrastructure. Its specific areas of conviction include AI infrastructure, orchestration, identity, data and security.
There is already a gap for those companies to fill. In a Team8 survey of 111 security leaders, 97% said their organizations had begun adopting AI agents and 80% were running them in production, while average confidence in their ability to secure AI stood at just 2.32 out of 5. The sample comes from Team8’s CISO Village Summit, so it should be read as an indicator from that community rather than a universal enterprise benchmark.
