Google alumni close $11M fund to back AI startups enterprises will pay for
What's the deal? BAG VenturesDealroom has a profile for this one. Try Dealroom →, founded by two GoogleDealroom has a profile for this one. Try Dealroom → alumni, has closed an $11.3 million venture fund to back early-stage AI startups. Founders Bontia Stewart, a former GoogleDealroom has a profile for this one. Try Dealroom → vice president, and Jackson Georges Jr., a former CapitalG partner, closed the fund after about two years of investing from it. The firm has already backed 10 companies, including software firm SXDDealroom has a profile for this one. Try Dealroom →, AI travel agent BizTripDealroom has a profile for this one. Try Dealroom →, and agentic reasoning platform NomadicDealroom has a profile for this one. Try Dealroom →.
The fund's focus: BAG invests in AI infrastructure, compute, physical and edge AI, security, governance, and vertical SaaS. Check sizes range from $100,000 to $500,000, and the team plans to deploy the rest of the fund over the next two years.
Who's backing it? BAG has more than 150 limited partners, including Google and operators from Nvidia, Amazon, and Snowflake. Stewart spent 17 years at Google, nearly a decade as a vice president, and sat on the board of Gradient VenturesDealroom has a profile for this one. Try Dealroom →, its early-stage AI fund. Georges, whom she met at Google, later became a partner at CapitalG, AlphabetDealroom has a profile for this one. Try Dealroom →'s growth fund.
What's the edge? The pair say their advantage is access. They launched BAG to "address the emerging AI divide between founders and operators," Georges said, promising warm introductions to potential customers and go-to-market advice. "Lots of firms have an operator network. We want to be one that's genuinely useful," he added.
Why now? Georges argues the "experimental sandbox" phase of enterprise AI is ending. "Enterprises are dialing in heavily on the unit economics right now," he said. "They aren't just paying for open-ended chatbots anymore; they are paying for deterministic solutions" that integrate into legacy workflows and execute real work, such as code reviews and parsing legal documents.
What could go wrong? Georges warns thin products won't survive as frontier labs ship more of their own. "If a startup is just a thin wrapper around a frontier model API, they're going to get wiped out," he said. BAG looks for teams that capture proprietary data and own "the customer lock-in to survive the next big model release."
The signal: BAG's thesis reflects a broader shift in how enterprises buy AI — from per-seat SaaS to paying for completed jobs and outcomes. Georges is betting on founders building deep into regulated workflows, where specialization and proprietary data offer durability that generic tools lack.
Read more: TechCrunch
Image credit: Generated with Gemini
