Center for Public Intelligence

Cover Block

PUBLIC

Attribute	Detail
Name	Center for Public Intelligence (CPI)
Tagline	A research lab building Public AI as digital public infrastructure for agentic government.
Headquarters	Washington, DC, USA
Founded	2025
Stage	Pre-Seed
Business Model	Other (Research Lab)
Industry	Defense / Govtech
Technology	AI / Machine Learning
Geography	Global / Remote-First
Growth Profile	Social Enterprise
Founding Team	Academic Spinout (University of Oxford)

Links

PUBLIC

• Website: https://publicintelligence.ai
• LinkedIn: https://uk.linkedin.com/company/publicintelligencecpi
• Instagram: https://www.instagram.com/publicintelligence.ai/

Executive Summary

PUBLIC The Center for Public Intelligence (CPI) is a conceptual research lab proposing to recast advanced AI as a public utility, a bet that government institutions will require new digital infrastructure to manage complexity and personalize services. Founded in 2025 by researchers from the University of Oxford and based in Washington, DC, CPI's core thesis is that AI should be architected as shared, open infrastructure,akin to roads or power grids,rather than remain a privately controlled technology [LinkedIn, September 2025]. Its proposed system, termed 'Agentic Government,' aims to move beyond simple automation toward institutions that can anticipate citizen needs [LinkedIn, September 2025].

The organization's differentiation lies in this philosophical and architectural positioning, advocating for 'Public Intelligence' as a universal right and a foundational component of modern governance [publicintelligence.ai]. The founding team's public profile is limited; available third-party data identifies Melissa Muñoz Suro as Managing Director and Lina Khayati as Director Marítimo, but detailed professional backgrounds or prior venture-scale operating experience are not documented in public sources [Prospeo Company Profile, 2025-2026].

As of early 2026, CPI shows no evidence of external venture funding, disclosed commercial contracts, or a traditional product SKU. Its financial metrics, including an estimated $2 million valuation and ~$599,000 in annual revenue, are third-party projections not tied to a financing event [Prospeo Company Profile, 2025-2026]. The next 12-18 months will be critical for validating whether this policy-focused research entity can translate its ambitious vision into tangible deployments, partnerships, or a sustainable funding model beyond academic grants.

Data Accuracy: YELLOW -- Core mission claims are self-published and corroborated across the organization's channels; financial and team details rely on a single third-party data vendor.

Taxonomy Snapshot

Axis	Classification
Stage	Pre-Seed
Business Model	Other
Industry / Vertical	Defense / Govtech
Technology Type	AI / Machine Learning
Geography	Global / Remote-First
Growth Profile	Social Enterprise
Founding Team	Academic Spinout

Company Overview

PUBLIC The Center for Public Intelligence (CPI) presents itself as an independent research lab, founded in 2025 by researchers at the University of Oxford and now headquartered in Washington, DC [LinkedIn, September 2025]. Its founding mission, as articulated in its own materials, is not to launch a commercial product but to create a new system of government, positioning AI as digital public infrastructure akin to roads or electricity [publicintelligence.ai]. This framing places the organization at the intersection of academic policy research and early-stage technology development, with a remote-first operational model.

Public milestones are sparse and self-published. The key documented event is the organization's formal introduction via a LinkedIn article in September 2025 [LinkedIn, September 2025]. No subsequent funding announcements, major partnership disclosures, or product launch events have been recorded in mainstream business or technology press. Third-party data estimates the organization employs between one and ten people [Prospeo, 2025-2026].

Data Accuracy: YELLOW -- Core founding story and mission are self-published; team size and financial estimates are from a single third-party vendor.

Product and Technology

MIXED The Center for Public Intelligence's public-facing product definition is conceptual, framed entirely as a research mission rather than a commercial offering. Its core proposition, articulated on its website and LinkedIn, is the development of 'Public AI' as a form of digital public infrastructure, analogous to roads or electricity grids, to enable what it terms 'Agentic Government' [publicintelligence.ai] [LinkedIn, September 2025]. This envisioned system would move beyond simple automation to allow governments to anticipate citizen needs and personalize public services at scale.

No specific product SKUs, technical specifications, or deployed software modules are described in public materials. The organization's output appears to be research and institutional design frameworks aimed at governments and public institutions. The technology stack is not publicly disclosed, though the focus on AI and agentic systems suggests a reliance on contemporary machine learning and large language model tooling (inferred from mission statements).

Data Accuracy: YELLOW -- Core mission statements are confirmed via the company's own website and social channels, but technical details, product roadmaps, and deployment evidence are absent from public sources.

Market Research

PUBLIC

The ambition to embed AI into the core of government operations arrives at a moment of acute pressure on public institutions to modernize, a demand driver that is more qualitative than quantitative but no less real for investors sizing the bet.

No third-party market sizing report specifically for "Agentic Government" or "Public AI as infrastructure" is available in the public record. The concept is too nascent and the company's own materials are conceptual, not commercial. For a sense of scale, investors can look to analogous markets. The global government AI market was valued at approximately $7.9 billion in 2023 and is projected to reach $71 billion by 2033, growing at a compound annual rate of 24.6% [Allied Market Research, 2024]. The digital government platforms segment, which includes citizen service portals and data integration layers, represents a multi-billion dollar subset of this. These figures, however, measure spending on software and services broadly, not the specific vision of AI as a public utility.

Government AI Market 2023 | 7.9 | $B
Projected Market 2033 | 71 | $B

The projected growth, while substantial, underscores a market still defining its boundaries; CPI's thesis depends on carving a new category within it, not capturing share from existing vendors.

Demand drivers are well-documented across public sector research. Aging IT infrastructure, rising citizen expectations for digital services, and workforce shortages are forcing modernization agendas [World Bank, 2023]. The tailwind for CPI's specific angle is a growing policy discourse, particularly in the EU and parts of Asia, around sovereign AI and digital public goods. This creates a receptive, if not yet purchasing, audience for the lab's research. The primary substitute market is not a competing product but the status quo: legacy government technology consultancies and large system integrators that deliver incremental automation, not systemic redesign.

Regulatory and macro forces cut both ways. Data sovereignty laws (e.g., GDPR, upcoming AI Acts) could complicate cross-border AI models but also bolster arguments for homegrown, public-sector-controlled infrastructure. Budget cycles and political risk remain the dominant constraints; a conceptual research lab without a clear procurement pathway faces a long adoption curve regardless of technical merit.

Data Accuracy: YELLOW -- Market sizing is from an analogous, broader sector report; demand drivers are cited from public institutions.

Competitive Landscape

MIXED The Center for Public Intelligence occupies a conceptual niche with few direct commercial rivals, competing instead against a diffuse set of incumbents, policy frameworks, and adjacent research initiatives.

The competitive map must be drawn from broader market categories.

CPI’s immediate competitive set is not other venture-backed startups but established incumbents in the government services and AI consulting space. These include large systems integrators like Accenture Federal Services and Deloitte’s Government & Public Services practice, which hold long-term contracts to modernize legacy IT and implement data analytics. Their advantage is entrenched relationships and procurement experience, but their offering is typically framed as efficiency and cost reduction, not a fundamental re-architecture of government intelligence. Adjacent substitutes include policy research institutes and think tanks, such as the Brookings Institution’s Center for Technology Innovation or the AI Now Institute, which produce influential frameworks and white papers on AI governance. These entities compete for mindshare and policy influence but do not claim to be building operational infrastructure. The most direct challengers are other nascent initiatives advocating for public-interest technology, such as The GovLab at NYU or the Digital Public Goods Alliance, though their focus is often on open data and digital public goods rather than specifically “agentic” AI systems.

Where CPI claims a defensible edge today is in its foundational intellectual positioning. The concepts of “Public AI as digital public infrastructure” and “Agentic Government” serve as a cohesive, academically-grounded narrative wedge. This edge is rooted in talent and ideas, sourced from its stated origin in University of Oxford research [LinkedIn, September 2025]. However, this edge is highly perishable. It is a first-mover advantage in framing, not in deployed technology, contractual relationships, or proprietary data. Without rapid translation into a demonstrable product, pilot, or policy adoption, this conceptual lead can be easily co-opted or overtaken by better-resourced entities that adopt the language.

The organization is most exposed in three areas. First, it lacks the capital and commercial track record of the large integrators who could decide to build or acquire similar capabilities. Second, it does not own a distribution channel into government procurement, a labyrinthine process that often sidelines pure research entities. Third, it faces competition from within the academic and non-profit world, where other labs may pursue similar missions with greater funding or established government partnerships. CPI’s current public footprint offers no counter to these exposures.

The most plausible 18-month competitive scenario hinges on validation. If CPI can secure a pilot project with a municipal or national government agency,even a non-commercial research partnership,it would solidify its position as a credible builder and attract talent and perhaps grant funding. The winner in this scenario would be CPI itself, carving out a specialized role as a translational lab between academia and applied public tech. The loser would be any similarly positioned, purely conceptual initiative that fails to secure such a foothold, remaining an interesting website without a path to impact. Conversely, if a major systems integrator or a well-funded nonprofit like OpenAI’s Public Benefit Corporation formally launches a “Public AI Infrastructure” practice within 18 months, they could rapidly absorb the market attention and resources CPI seeks, leaving it marginalized.

Data Accuracy: YELLOW - Competitive analysis is inferred from market categories and public positioning due to a lack of named, directly comparable competitors in sources.

Opportunity

PUBLIC The Center for Public Intelligence's opportunity rests on the potential to define and own the foundational software layer for a new era of proactive, AI-native governance, a category that currently exists only in concept.

The headline opportunity is to become the default provider of the digital public infrastructure for agentic government, analogous to what AWS became for cloud computing or Stripe for payments. The company's core thesis, that AI should be a shared public resource akin to roads or electricity, positions it to capture the first-mover advantage in a market that is currently a collection of disparate, reactive point solutions [publicintelligence.ai]. The outcome is reachable not because of current traction, but because the underlying premise,that governments globally are under pressure to modernize and personalize services,is widely accepted. If CPI can translate its research into a licensable platform or standard, it could embed itself into the procurement cycles of national and municipal governments as the reference architecture for public AI.

Growth would likely follow one of several distinct, non-linear paths. The scenarios below outline concrete routes to scale, each requiring a specific, plausible catalyst.

Scenario	What happens	Catalyst	Why it's plausible
Oxford Policy Standard	The lab's academic origins lead to its frameworks being adopted as a de facto policy standard for AI in public services, first in the UK/EU and then globally.	Publication of a seminal white paper or research partnership with a major intergovernmental body (e.g., UN, OECD) that codifies its "Agentic Government" principles [LinkedIn, September 2025].	The University of Oxford pedigree provides a credible platform for policy influence, and the vacuum of clear standards for public AI creates an opening for a principled, academic approach to gain traction.
Pilot-to-Platform Land Grab	CPI wins a single, high-profile pilot with a forward-leaning city or national agency (e.g., Estonia, Singapore, a U.S. "smart city"), then uses the deployment as a reference case to sell a standardized SaaS platform.	Securing a pilot contract, likely non-dilutive via a grant or innovation fund, to implement a specific use case like personalized education or welfare services [LinkedIn].	Governments increasingly run innovation sprints and sandboxes for AI; a research lab is a structurally lower-risk partner for an experimental pilot than a venture-backed commercial vendor.
Infrastructure Spin-out	The research crystallizes into a distinct, venture-backable software company (a "CPI Labs" spin-out) focused on the underlying agentic orchestration layer, which is then adopted by systems integrators and govtech primes.	A strategic investment from a tier-1 venture firm or a government-focused fund to commercialize the IP, separating the research mission from a for-profit product roadmap.	The historical pattern of foundational tech (e.g., internet protocols, cryptography) emerging from research institutions before commercial exploitation is well-established.

What compounding looks like is a classic standards-and-ecosystem flywheel. An initial policy win or pilot deployment would generate proprietary data on public service interactions and institutional workflows. This dataset, unique to a government context, would improve the performance and specificity of CPI's AI models, making the platform more valuable for the next jurisdiction. Furthermore, early adoption by reputable institutions would attract developer and partner ecosystems to build on the platform, increasing switching costs and entrenching CPI's architecture as the lingua franca for public-sector AI applications. The flywheel is currently at stage zero; the cited evidence shows the conceptual blueprint but not yet the motion [publicintelligence.ai].

The size of the win can be framed by looking at the valuation of companies that have successfully built foundational infrastructure in adjacent sectors. For instance, Palantir Technologies, which provides operating systems for government and enterprise data, reached a public market capitalization exceeding $40 billion. While CPI's scope is different, the comparable illustrates the premium awarded to a platform that becomes deeply embedded in critical government workflows. If the "Pilot-to-Platform" scenario plays out, CPI could aim to capture a segment of the global govtech AI market. A conservative scenario valuation might look to acquisition multiples for specialized SaaS platforms in regulated industries, which often trade at 10-15x revenue. Applying that to Prospeo's estimated current revenue (approximately $600k) is not meaningful for a pre-product entity. The more relevant figure is the potential addressable revenue if it captured even a single-digit percentage of a major government's digital transformation budget, which can range from hundreds of millions to billions annually. A successful outcome here is not about incremental efficiency gains but about defining and supplying the core infrastructure for a trillion-dollar shift in how states operate.

Data Accuracy: YELLOW -- The opportunity analysis is derived from the company's stated mission and academic positioning, which are confirmed via its own channels. The growth scenarios and comparables are plausible extrapolations, not statements of fact. The estimated revenue figure is from a single third-party source.

Sources

PUBLIC

1. [LinkedIn, September 2025] Introducing the Center for Public Intelligence (CPI) | https://www.linkedin.com/pulse/introducing-center-public-intelligence-cpi-publicintelligencecpi-dj1qe

2. [publicintelligence.ai, current] CPI: Home | https://publicintelligence.ai

3. [Prospeo Company Profile, 2025-2026] Center for Public Intelligence (CPI) Revenue, Funding & Valuation | https://prospeo.io/c/center-for-public-intelligence-cpi-revenue

4. [Allied Market Research, 2024] Government AI Market | https://www.alliedmarketresearch.com/government-artificial-intelligence-market-A31766

5. [World Bank, 2023] Digital Government | https://www.worldbank.org/en/topic/digitalgovernment

6. [LinkedIn] Center for Public Intelligence (CPI) | https://uk.linkedin.com/company/publicintelligencecpi