I said it back in 2022 and got ratio’d for it: PFPs weren’t a bubble; they were an early prototype for verifiable digital identity. The market just wasn’t ready for that framing yet.
Key Takeaway
As AI agents begin transacting, communicating, and making decisions on behalf of humans across the web, the absence of a robust digital identity layer – one that can verify both human and agent principals – creates fundamental risks for trust, accountability, and the attribution of actions in digital systems.
People treated PFPs like JPEG Beanie Babies – flip for profit, flex on timeline, dump when boring. But the tech was always deeper than that: on-chain ownership, unique traits as reputation signals, communities built around shared aesthetics. Bored Apes and CryptoPunks showed we could recognise each other instantly in a sea of anonymity. That instinct – trust through a recognisable face – never went away. It just went dormant for a bit.
Key Takeaways
- The Numbers Don’t Lie
- The Real Catalyst: AI Agents
- The Script Flips
- My Prediction
The Numbers Don’t Lie
Fast-forward to now (February 2026): NFT volumes are climbing again. Q3 2025 alone saw $1.58 billion in trades and a record 18.1 million sales (DappRadar numbers). Polymarket odds for a real NFT comeback in 2026 hit 65% ATH in January.
This isn’t 2021 redux. Speculation is fading; utility is rising – IP, ticketing, loyalty, genuine provenance. The use cases that actually hold up are starting to separate from the ones that never made sense.
The Real Catalyst: AI Agents
Autonomous economic actors are here: trading, creating content, negotiating, governing DAOs, managing personal finances, curating feeds. As they flood on-chain ecosystems, they face the exact same problem humans faced before them. How do you build trust when everyone is just anonymous code?
Humans trust faces. We read expressions; we spot fakes from micro-cues we can’t even name consciously. An AI agent with a recognisable, verifiable PFP inherits that same signal. It’s not decoration – it’s KYA (Know Your Agent).
A CryptoPunk on your trading bot says “proven lineage, elite community backing, track record attached.” A Bored Ape signals membership in a high-reputation network. Agents will adopt PFPs because reputation primitives beat anonymous wallets every single time, and that’s not going to change.
The Script Flips
In 2021, humans bought PFPs for status. In 2026–2028, agents buy – or get assigned – PFPs for functionality. Demand will come from non-human users who need to signal reliability fast, in contexts where there’s no time for a reputation check.
Platforms will prioritise verified-agent PFPs in their interfaces. Social protocols might default to NFT avatars for bots. The most liquid, highest-floor collections could end up agent-owned precisely because they pay more utility value than we ever paid in flex dollars. That’s a strange outcome to sit with, but the logic is sound.
My Prediction
I think the most valuable PFP collection in 2028 won’t have a single human owner left. The agents are coming for your profile pic – and they’ll bid higher, because for them identity isn’t optional. It’s infrastructure.
PFPs were early. The world just caught up. Now watch them become boringly essential.
Further reading: How AI Agents Are Changing Business Operations, AI Agents vs Chatbots: Why the Difference Matters, Building Autonomous Workflows with AI Agents.
Frequently Asked Questions
Why do AI agents create a digital identity problem?
When AI agents act autonomously on your behalf – sending emails, making purchases, booking appointments – the receiving party has no reliable way to verify whether a human or an AI made the decision, whether the AI was actually authorised to act, or who is accountable if the action causes harm. Existing identity systems were designed for human actors, not AI agents operating with delegated authority.
What would a robust AI agent identity framework look like?
A robust framework would include: cryptographic delegation chains showing which human principal authorised the agent, capability attestations describing what the agent is permitted to do, tamper-resistant and attributable audit logs, revocation mechanisms that can terminate agent authority, and standardised identity tokens that third-party systems can verify independently.
How does AI agent identity relate to PFPs and digital identity NFTs?
Early experiments with PFPs as verifiable digital identity signals were primitives pointing toward something deeper – on-chain identity infrastructure. The underlying challenge is the same: creating portable, verifiable, privacy-preserving identity attestations. The difference is in the stakes. When AI agents hold economic authority rather than just aesthetic identity, the infrastructure needs to meet financial-grade security requirements.
AI Agents Are Coming for Your Digital Identity
