<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0">
	<channel>
		<title>Yaman Can · Blog</title>
		<link>https://yamancan.com/blog</link>
		<description>Products for AI agents, applied cryptography, and notes from the build.</description>
		<language>en</language>
		<item>
			<title>The cluster is the result. xAI is the exception.</title>
			<link>https://yamancan.com/blog/agree-with-everything</link>
			<guid isPermaLink="true">https://yamancan.com/blog/agree-with-everything</guid>
			<pubDate>Sat, 29 Aug 2026 00:00:00 GMT</pubDate>
			<description>In a 22-model T=0.1 political-questionnaire cohort, all 20 non-xAI models across nine US- and China-headquartered providers answered from the same equality–liberty region. Release drift was family-level, temperature had no common direction, and the country comparison depended on one provider.</description>
		</item>
		<item>
			<title>An image is not a sentence</title>
			<link>https://yamancan.com/blog/vson-predicate-logic</link>
			<guid isPermaLink="true">https://yamancan.com/blog/vson-predicate-logic</guid>
			<pubDate>Thu, 06 Aug 2026 00:00:00 GMT</pubDate>
			<description>From an analytical-philosophy obsession to VSON: a notation where every claim about an image declares every argument it takes, viewpoint included.</description>
		</item>
		<item>
			<title>Trick the agent all you want. The vault does not care.</title>
			<link>https://yamancan.com/blog/trick-the-agent</link>
			<guid isPermaLink="true">https://yamancan.com/blog/trick-the-agent</guid>
			<pubDate>Thu, 06 Aug 2026 00:00:00 GMT</pubDate>
			<description>Assume prompt injection will happen. Maestro puts ten on-chain checks between a hijacked agent and the money, and is honest about what an attacker still gets.</description>
		</item>
		<item>
			<title>Three agents, one signature, none holds the key</title>
			<link>https://yamancan.com/blog/three-agents-one-signature</link>
			<guid isPermaLink="true">https://yamancan.com/blog/three-agents-one-signature</guid>
			<pubDate>Thu, 06 Aug 2026 00:00:00 GMT</pubDate>
			<description>How Chorus runs an AI-agent committee on FROST threshold signatures and ERC-7710 delegation: no key to steal, no authority that was not explicitly given.</description>
		</item>
		<item>
			<title>App islands: an edge you can rent is not an edge</title>
			<link>https://yamancan.com/blog/app-islands</link>
			<guid isPermaLink="true">https://yamancan.com/blog/app-islands</guid>
			<pubDate>Thu, 06 Aug 2026 00:00:00 GMT</pubDate>
			<description>AI collapsed the cost of building your own internal tools. Notes on skipping the CRM in San Francisco, and why the next moat is the internal codebase.</description>
		</item>
		<item>
			<title>Why we ship products from cryptography papers</title>
			<link>https://yamancan.com/blog/ship-from-papers</link>
			<guid isPermaLink="true">https://yamancan.com/blog/ship-from-papers</guid>
			<pubDate>Wed, 20 May 2026 00:00:00 GMT</pubDate>
			<description>There is a long gap between a zero-knowledge paper and something a person can click. Notes on how we close it.</description>
		</item>
		<item>
			<title>Hiding zero-knowledge latency is a UX problem</title>
			<link>https://yamancan.com/blog/hiding-zk-latency</link>
			<guid isPermaLink="true">https://yamancan.com/blog/hiding-zk-latency</guid>
			<pubDate>Wed, 08 Apr 2026 00:00:00 GMT</pubDate>
			<description>Proofs take seconds. Players will not wait seconds. What designing Mina Mastermind taught me about masking compute.</description>
		</item>
		<item>
			<title>AI agents need wallets. They also need a kill switch.</title>
			<link>https://yamancan.com/blog/agent-wallets-guardrails</link>
			<guid isPermaLink="true">https://yamancan.com/blog/agent-wallets-guardrails</guid>
			<pubDate>Sun, 15 Feb 2026 00:00:00 GMT</pubDate>
			<description>Off-chain guardrails are bypassable middleware. The case for putting agent spending policy on-chain.</description>
		</item>
	</channel>
</rss>