All takes are our summaries. Tap View on Xfor the analyst's original words.
In a world saturated with AI agents, Altman's Worldcoin identity project becomes essential infrastructure — you need a provably-human layer. Fernando frames identity-for-AI as a category hiding in plain sight: when 'more things look like people than people do', the iris-scan primitive becomes the on-ramp for every other consumer product that needs to distinguish humans from bots.
Thesis: compute becomes a commodity, like oil. Supply-constrained today, but heading toward standardization. Like oil, it needs market infrastructure — futures, storage/logistics, price discovery, hedging instruments. First movers are the cloud operators; the real prize is the exchange layer that gets built atop them. The venture opportunity is backing that layer, not the underlying chips or data centers.
Kevin examined AI agent investment opportunities and identified where moats actually exist. The sector's real defensibility lies not in engineering patterns—which open source reimplements in weeks—but in proprietary trajectory data from execution, integration depth with customer systems, and evaluation infrastructure. Companies like Harvey ($190M ARR), Sierra ($150M+ ARR), and Cursor ($2B ARR) compound advantages through data flywheels, while Meta's $2 billion Manus acquisition signaled that 147 trillion tokens of execution data across 80 million VM sessions justifies premium valuations where framework elegance and generic tooling offer no moat.
Robbie argues the agentic commerce thesis is inverted: while the agentic economy will be massive, most agents won't transact autonomously. Commercial agents (95%+ of agentic deployment) embedded in SaaS won't spend money—they'll research, review, or generate output. Consumer agents will remain orchestrators requesting authorization, not independent economic actors. Only bottom-up agents outside organizational control genuinely need granular, autonomous payments, where blockchains' permissionlessness beats card networks' compliance friction. The real bottleneck isn't payment rails but regulatory frameworks and legal structures enabling autonomous decision-making—a protocol upgrade can't solve that. Most agentic economy activity gets billed monthly, not settled per transaction.
Donovan argues AI agents on blockchains remain mostly a meme: x402 onchain transactions peaked in November 2025 then collapsed, with merchants offering only speculation plays rather than useful services. Three binding constraints prevent growth: discovery (no registry of x402-enabled services), identity (no way to verify unknown wallets), and reputation (no chargeback mechanisms). The missing layer is an agentic PageRank combining onchain volume, attestation reviews, and completion rates—whoever builds it could own the agentic economy's monetization funnel, potentially larger than Google's AdWords.
Sam argues Galaxy's recent 30% gain understates its potential given newly approved 830 MW at Helios (doubling approved capacity to 1.6 GW) layered onto an already-cheap valuation. The 1.6 GW scenario implies ~$80/share equity value at full build and ~$40/share present value, combined with Galaxy's Digital Assets segment (60% of base case, supported by CLARITY crypto legislation and on-chain innovations like tokenized equity and commercial paper issuance) points to total valuation above $80/share versus Friday's $31.90 close. Execution on contracting the new power tranche and visible construction progress remain critical catalysts.