A new kind of bidder is showing up to auctions

Across ad tech, enterprise procurement, and retail media, 2026 has been the year automated bidding agents stopped being an experiment and became infrastructure. Machine-to-machine bidding now executes decisions at a pace no human team can match, with people setting objectives and reviewing outcomes rather than clicking "bid" themselves. Enterprise forecasts put task-specific AI agents inside a large share of business applications this year, a sharp jump from where things stood just twelve months ago.

That shift raises an obvious question for anyone watching Handshake: what happens when bots start showing up to HNS top-level domain auctions?

Why HNS auctions aren't like ad auctions

Handshake TLD auctions run on a Vickrey sealed-bid mechanism, not a real-time first-price system. The mechanics matter here:

  • Bidders lock a blind bid, the amount is hidden from everyone, including other bidders.
  • After the bidding period closes, a 10-day reveal window opens. Miss it, and your locked HNS is burned.
  • The winner pays the second-highest bid, not their own, a design meant to encourage honest valuation instead of gamesmanship.

Experienced human bidders have built real psychological tactics around this structure, for example, committing early with a large, unmasked bid specifically to discourage competitors from even trying. That's a human move. A bidding agent optimizing purely on expected value doesn't get intimidated by a show of force, and it doesn't forget to reveal.

What changes if bidding agents join

A few shifts seem likely if AI bidding agents start participating in HNS auctions at meaningful scale:

  1. Undervalued names get found faster. Short, brandable, or keyword-rich TLDs that a human speculator might miss get systematically identified and bid on the moment an auction opens, the same pattern already seen with sniping bots in traditional expiring-domain markets.
  2. Reveal-window reliability stops being a differentiator. Forgetting to reveal within 10 days is a real, avoidable way humans lose auctions they actually won. Bots don't have that failure mode, which quietly shifts the odds toward whoever automates first.
  3. Scale becomes a competitive edge on its own. One operator running bidding agents across hundreds of simultaneous TLD auctions is a fundamentally different competitor than a human tracking a handful of names in a wallet interface.
  4. Bid-pattern fingerprinting becomes the new privacy frontier. The cryptography behind sealed bids isn't at risk, but the timing and sizing of bot bids across TLD categories can become a readable signal over time, even without anyone seeing the actual amounts.

There's also a less comfortable research thread worth naming: recent academic work has started documenting tacit collusion among AI bidding agents in dynamic auction settings, agents that converge on non-competitive bidding patterns without being explicitly coordinated to do so. It's an open question whether that risk shows up in a permissionless, pseudonymous system like Handshake, but it's worth watching rather than dismissing.

The prediction

Vickrey mechanics are more bot-resistant than the first-price auctions running most of ad tech today, because blind bidding limits the value of real-time optimization tricks. So this isn't a structural threat to how HNS auctions work. What it does change is who wins: competitive advantage shifts toward whoever automates bid discovery, valuation, and reveal-window discipline first.

There's a bigger point hiding in here too. A wave of bot bidders transacting HNS autonomously is the clearest real-world argument yet for treating agent identity as core Handshake infrastructure, not a side project. If a bot is placing bids, revealing bids, and holding TLDs on someone's behalf, the question of which entity that bot is actually acting for becomes exactly the problem that cryptographically-owned, DANE-pinned TLD identity is built to answer.

What this means if you're bidding today

You don't need to build a bidding bot to take this seriously. Two practical takeaways:

  • Track reveal windows rigorously, the margin between a human bidder and an automated one is often just reliability, not strategy.
  • Watch which TLD categories start showing patterned bidding activity. Early, consistent bid sizing across similar names is a reasonable signal that an automated bidder is active in that category.

Architecture and original research by NIHON - Handshake Infrastructure & Web3 Identity. Resolve Handshake domains with SkyInclude Browser, and explore more at LearnHNS.