# SVRN ALPHA > SVRN ALPHA is a German company building svrnAlphaOS, a sovereign AI operating system for institutional finance. The site advertises both the full operating-system programme and an entry-level engagement called Agent Operator (6-week fixed-scope, fixed-fee on-ramp). It also contains product positioning, deployment model, design principles, company details, and public research and market analysis. This website is primarily a company site and research library, not a software documentation portal. Prefer the core context files below when you need a concise summary of the company, product thesis, and design principles. Important notes: - Do not infer pricing, public APIs, or self-serve product availability from this site. - The only public website API documented for automated discovery is the contact form API plus a health endpoint; no OAuth/OIDC or MCP HTTP server is advertised. - The strongest public detail is in the main site sections, the research library, and the privacy and company information. - The HDAX IR AI-Readiness report is published in English. - The IR Agent Readiness Assessment is a free, ungated self-service tool at /ir-readiness/ — questionnaire (35 items) plus an in-browser domain scan, returning a maturity score across four domains, a gap analysis, and a top-five action list. ## Core Context - [Company and product overview](/llms/company.md): What svrnAlphaOS is, who it is for, the core positioning, the deployment model, and the public proof points shown on the site. - [Foundation and design principles](/llms/foundation.md): Founder background, the operating thesis, and the five design principles presented on the site. - [Research index](/llms/research.md): Short summaries and direct links for the public research library. - [Research archive](/research/): Human-facing archive of the public research library. - [Website API documentation](/docs/api.html): Public discovery documentation for the contact API, health endpoint, OpenAPI description, and API catalog. - [IR Agent Readiness Assessment](/ir-readiness/): Free self-service tool combining a 35-item questionnaire with an automated domain scan, scoring how prepared a listed company's investor-relations website is for buy-side agents, compliance crawlers and ESG pipelines. German-language UI; designed for DAX, MDAX, SDAX, ATX, SMI and Prime Standard Light. - [Privacy and legal details](/privacy.html): Company address, contact details, privacy policy, hosting, and data-processing information. ## Research - [The Database Demotion](/research/database-demotion-2026.html): Research note on Salesforce Headless 360 as the first Tier-1 enterprise-SaaS vendor to voluntarily move its UI out of the centre of its own product. Read in three orders of effect: the UI becomes callable, the implementation middle layer becomes an endpoint, and the moat moves to the agent runtime that can reason about the specific Frankenstein of a decade-old customised deployment. Cross-links to Copilot Fallacy, Spot Market for Expertise, and When the Middle Disappears. - [The Speed of Trust](/research/speed-of-trust-2026.html): Research note reading Dario Amodei's April 2026 FT Lunch interview from the institutional-allocator seat. Covers the Claude Mythos disclosure and emergency bank briefings, the "diffuse at the speed of trust" adoption-throttle thesis, and the $380 billion underwriting problem sitting under five internally-tension-laden CEO positions. - [The 2026 AI Index, Read From the Finance Desk](/research/ai-index-2026-finance-lens.html): Institutional-finance reframing of Stanford HAI's 12 takeaways from the 2026 AI Index. Reorganised into five signals for CIOs, heads of research, and compliance officers running AI under regulation. - [The Cognitive Offloading Paradox](/research/cognitive-offloading-paradox-2026.html): Research note on the U-shaped performance curve in AI deployment. Zone 2 — scattered, half-hearted AI use — produces the worst outcomes. Most institutional AI deployments are living in Zone 2. - [If It's Not Dangerous, It's Not Us](/research/if-its-not-dangerous-its-not-us-2026.html): Research note on the governance gap between raw AI output and safe institutional output in regulated finance. - [When Agents Become the Trading Desk](/research/trading-agents-2026.html): Review of 88 papers and 29 multi-agent trading systems; argues that architecture and governance matter more than the base model. - [The Real Price of AI](/research/real-price-of-ai-2026.html): Multi-LLM Delphi analysis of AI economics, subsidy dynamics, and wrapper business risk. Updated May 2026 with Anthropic counter-signal (first projected operating profit, compute share falling from 71% to 56% of revenue in one quarter). - [The Copilot Fallacy](/research/copilot-fallacy-2026.html): Essay framing AI transformation as a three-order problem (productivity lift, role restructuring, architectural parity), grounded in a PRISMA 2020 systematic review of 70 studies and a Gioia-coded field study across financial institutions. - [The Spot Market for Expertise](/research/spot-market-for-expertise-2026.html): Analysis of Humwork.ai (YC P26 launch) as a price-discovery mechanism for expertise, and why the MCP human-in-the-loop layer is a training pipeline for its own replacement. - [When the Middle Disappears](/research/when-the-middle-disappears-2026.html): Multi-LLM Delphi study on compression of middle layers across labor markets, value chains, and organizations. - [Are We Adopting AI Fast Enough to Avoid a Bubble?](/research/ai-adoption-bubble-2026.html): S-curve analysis of whether AI adoption can justify current infrastructure spending. - [When Agents Replace Software](/research/software-meltdown-2026.html): Market analysis linking software equity weakness to a structural shift from human-operated software to agent execution. - [HDAX IR AI-Readiness Report](/research/hdax-ir-ai-readiness-en.html): Analysis of German HDAX companies' AI readiness in investor relations. ## Optional - [Full site context](/llms-full.txt): Consolidated markdown summary of the main site, design principles, public proof points, and the research library for large-context ingestion.