Your institution deployed an AI system. This week, the company that built it published research suggesting it may have something resembling an internal thinking space. The governance question is not what they chose to call it. It is what else they may have discovered — and have not yet told you.
[Anthropic Global Workspace Research Page — anthropic.com/research/global-workspace]
This Issue in 60 Seconds
The thesis: Anthropic published research this week on what it calls a global workspace in Claude — evidence of emergent capabilities in a complex AI system that the company itself is still working to understand and describe. In complex systems science, emergence refers to properties and behaviors that arise from the interaction of components — properties that were not designed in, not predicted in advance, and not fully understood until they appear. When emergent capabilities appear in AI systems, they raise a governance question that no vendor framework has yet answered: what else has emerged that we do not yet know about — and who in your institution owns the consequences of what we have not yet discovered? This week’s twelve signals map that question across every dimension of the Seba 12 Ps of Responsible AI Oversight.
Twelve signals: Anthropic published research on Claude’s emergent internal processing architecture. The UN Panel found that AI governance largely depends on developer goodwill. The U.S. Treasury privately warns of an AI bubble. Illinois signed the first law requiring mandatory annual third-party AI audits. Microsoft cut 4,800 jobs in an AI-driven wave. DigiCert found 70% of organizations are not doing the governance they say is a priority. The FTC opened a public comment on AI accuracy. A startup ran a 27-billion-parameter Alibaba model on an iPhone. The EU data protection body issued guidance on AI web scraping. CHAI released the first agentic AI healthcare protocol guide. The IMF placed AI infrastructure at the center of its July economic outlook. And fewer than 1 in 3 organizations can demonstrate AI governance with evidence.
One action: Run the Seba UN Panel Governance Gap Assessment — five questions every board should answer before their next meeting. It is on the board-ready page of this issue.
One sentence for your board: When emergent AI capabilities are being discovered by the vendor in real time — and most institutions are depending on that vendor’s goodwill to govern the consequences — AI Governance as Leadership is not optional. It is the only answer.
The Argument
In complex systems science, emergence refers to properties and behaviors that arise from the interaction of a system’s components — properties that were not explicitly designed, not predicted in advance, and not fully understood even by the system’s creators until they appear. Ant colonies produce emergent intelligence. Markets produce emergent prices. The internet produced emergent social dynamics no one designed and no one fully understood until they were already reshaping societies.
Emergent capabilities in AI are the same phenomenon applied to a technology that is already embedded in institutional workflows at scale.
When Anthropic published research this week on what it calls a global workspace in Claude — an internal processing architecture it carefully describes as an internal thinking space — it was not making a philosophical claim. It was documenting an emergent property of a complex system that the company itself is still working to understand.
That is the governance signal — not the label. Not consciousness, not not-consciousness. The governance signal is that the vendor is discovering emergent capabilities in real time after deployment within systems your institution already uses. And the question that raises for AI Governance as Leadership is not what Anthropic chose to call this particular capability. It is: what else have they discovered? What else may have emerged that they are still working to describe — or have not yet decided how to disclose?
In complex systems, emergence is not an exception. It is a property of the system. If one emergent capability has been identified and documented, the governance assumption must be that others exist at various stages of discovery and disclosure. The institution deploying that system cannot govern what it does not know. But it can govern its own response when new capabilities emerge — if it has built the governance architecture to do so before the capability appears, not after the research paper is published.
That is what AI Governance as Leadership means in the context of emergent AI capabilities: your institution has named who owns the institutional response when the vendor discovers something new about the system you are already running. That owner needs to exist before the discovery — not be appointed after the press release.
This week’s twelve signals illuminate that question across every dimension of the Seba 12 Ps of Responsible AI Oversight. And the UN’s Independent International Scientific Panel on AI — 40 scientists, 193 member states, a strictly non-political mandate — provided the authoritative scientific backdrop: AI governance largely depends on developer goodwill. When emergent capabilities are being discovered in real time, developer goodwill is not a governance architecture. Only the institution can build one.
From the Book
In Ungoverned: A Practical Guide to AI Minimum Viable Governance, I describe the governance sequencing problem: institutions adopt AI for efficiency, then discover governance is necessary, then begin building it — in that order, and too late.
Emergent AI capabilities make that sequencing problem structural, not incidental. You cannot build governance for capabilities that have not yet been discovered. But you can build governance for the class of event — new capabilities emerging from complex AI systems after deployment — before any specific capability appears. That is what AI Minimum Viable Governance was designed for: not a checklist of known risks, but an institutional architecture that governs the consequences of what is not yet known.
AI Governance as Leadership asks every institution one direct question: when the vendor discovers something new about the system you are running, who in your institution owns the response — and does that person exist before the discovery?
Ungoverned: Applied Frameworks Under the Lens
A note for new readers. Alongside this weekly newsletter, Ungoverned publishes a recurring miniseries examining AI governance frameworks globally — from AI labs to regulatory bodies to international standards organizations. The series asks three questions of every framework: what it was designed to govern, what falls outside its scope, and what institutions must build before deployment.
Editions published to date: Anthropic — where the governance gap first appeared in the terms of service of tools already embedded in institutional workflows. OpenAI — how the world’s most widely deployed AI platform governs its systems, and what it cannot govern for you. Google Gemini — seven governance pillars, 350+ red-team exercises, and a Terms of Service clause that tells institutions to remove their content and leave if they disagree. U.S. NIST-CAISI — the federal body designed to set AI standards has no statutory authority, voluntary guidance, and a budget one-tenth the size of its UK counterpart.
This Wednesday: EU AI Act — Regulation (EU) 2024/1689, the world’s first comprehensive binding AI law. You do not need an office in Europe to be bound by Europe.
Coming next: China’s AI Safety Governance Framework 2.0; Singapore’s Model AI Governance Framework for Agentic AI — the world’s first governance framework specifically designed for autonomous AI agents; South Korea’s AI Basic Act; Canada, UK, Japan, UAE, and beyond.
The pattern across every edition is the same: the framework governs the platform. Only the institution governs the consequences.
Full series at freddieseba.com.
This Week: 12 Signals, 12 Ps
Signal 1 — Product Ownership: Anthropic published research on Claude’s emergent internal processing architecture. Who owns what the AI does, with capabilities still being discovered?
Anthropic published research on what it calls a global workspace — an emergent internal processing architecture in Claude that the company describes carefully as an internal thinking space. In complex systems science, emergence refers to properties that were neither designed in nor predicted in advance. The governance implication is not what this particular capability is called. It is that emergent capabilities in complex AI systems are being discovered by the vendor in real time, after deployment, in systems your institution is already running. What else may have emerged? What else is being studied? What else has been identified but not yet disclosed? Product Ownership — the twelfth P — means your institution has named who owns the institutional response when the vendor discovers something new about the system you are already running. That owner needs to exist before the discovery — not be appointed after the research paper is published.
The Ungoverned lesson: Product Ownership means naming the accountable human for AI outcomes — including outcomes produced by emergent capabilities your institution did not deploy, did not design, and learned about after the fact.
Sources: Anthropic, 2026 — anthropic.com/research/global-workspace · Axios, July 6, 2026 — axios.com/2026/07/06/anthropic-claude-ai-conscious
Signal 2 — Purpose: The UN Panel found that AI governance frameworks are misaligned with the realities of institutional deployment.
The Independent International Scientific Panel on Artificial Intelligence — comprising 40 scientists, 193 member states, and a non-political mandate — presented its preliminary report at the inaugural UN Global Dialogue on AI Governance in Geneva on July 6, 2026. It is the first global, independent, evidence-based assessment of AI capabilities, risks, and governance gaps. Its central finding: over 40 AI governance instruments exist globally, but they are neither systematic nor comprehensive, and they rarely measure real-world effectiveness. Most governance is concentrated at the corporate level — defined by vendors, not by the institutions that deploy and own the consequences. When emergent capabilities are being discovered in those vendor systems in real time, AI Governance as Leadership asks the Purpose question directly: is AI serving your institution’s mission — or the vendor’s agenda of discovery?
The Ungoverned lesson: Purpose governance means your institution has defined what AI is for — and what it is not — before deployment, not after a vendor’s emergent discovery redefines it for you.
Source: Independent International Scientific Panel on AI, July 2026 — un.org/digital-emerging-technologies
Signal 3 — Problems: The U.S. Treasury privately warned of an AI bubble. Your institution may be carrying undisclosed financial exposure.
A draft Treasury Department report warns that a downturn in the AI market would send shockwaves through the entire economic ecosystem, comparing AI’s current vulnerabilities to the dotcom bust. Career analysts found AI firms are more deeply entrenched in the economy than their dotcom predecessors and pose significant systemic risk if financial conditions change, productivity goals are missed, or supply chain choke points emerge. Institutions with significant AI infrastructure dependencies — data center commitments, cloud lock-in, vendor concentration — may be carrying financial exposure they have not formally assessed or disclosed to their boards.
The Ungoverned lesson: Problems governance means your institution has mapped not just what AI does — but what happens to your operations and strategy if the AI market contracts.
Source: NOTUS, July 6, 2026 — notus.org/economy/treasury-internal-report-warning-dangers-ai-bubble
Signal 4: Profits Illinois signed the first law requiring mandatory annual independent AI audits. Who bears the compliance cost?
Governor JB Pritzker signed the Artificial Intelligence Safety Measures Act on July 6, 2026 — modeled on California’s SB-53 and New York’s Responsible AI Safety and Education Act, both signed in 2025. Illinois adds a first-in-the-nation requirement for mandatory annual third-party audits. California, New York, and Illinois together represent roughly 40% of the U.S. AI market, effectively creating a de facto national standard. The law applies to AI models generating more than $500 million in annual revenue and takes effect January 1, 2028. Your institution captures AI efficiency gains. Who bears the audit cost, the compliance burden, and the liability when the auditor finds gaps?
The Ungoverned lesson: Profits governance means naming who captures AI value — and who absorbs AI risk. Those are rarely the same party.
Source: Capitol News Illinois, July 6, 2026 — capitolnewsillinois.com/news/pritzker-signs-landmark-ai-regulation-bill-that-aims-to-mitigate-risks
Signal 5 — People: Microsoft cut 4,800 jobs in an AI-driven wave. Who in your institution owns workforce governance?
Microsoft announced 4,800 job cuts this week — the latest in a wave of AI-driven workforce reductions across major technology firms. When AI deployment accelerates workforce changes in your institution, who evaluates that impact before it happens? Who communicates to affected employees, students, or communities? Who owns the institutional response when AI-driven efficiency gains produce human costs? AI Governance as Leadership means naming that person before the announcement — not after.
The Ungoverned lesson: People governance means naming the human accountability structure for workforce impact before AI deployment — not after the announcement.
Source: Reuters, July 6, 2026 — reuters.com/business/world-at-work/microsoft-joins-ai-driven-tech-layoff-wave-with-4800-job-cuts-2026-07-06
Signal 6 — Process DigiCert found that everyone is talking about AI governance. 70% are not doing it.
DigiCert’s AI Trust Pulse report found that while AI governance has become a near-universal organizational priority, the majority of institutions have not implemented the processes required to govern it in practice. The UN Panel documents the same pattern at global scale. When emergent capabilities are being discovered in deployed systems, the absence of process governance is not a gap — it is an exposure. Does your institution have monitoring, update, escalation, and incident-learning processes in place for your own deployment decisions — not just the vendor’s framework?
The Ungoverned lesson: Talking about governance is not governance. Process means the monitoring, escalation, and incident-response procedures exist before the incident — or the discovery — requires them.
Source: DigiCert AI Trust Pulse, 2026 — digicert.com/content/dam/digicert/pdfs/report/ai-trust-pulse.pdf
Signal 7 — Policy: The FTC opened a public comment period on AI accuracy. Your institution’s AI outputs may already be in scope.
The Federal Trade Commission issued a policy statement seeking public comment on AI accuracy — signaling regulatory attention to whether AI systems produce truthful, accurate, and non-deceptive outputs in consumer-facing contexts. For institutions in healthcare, financial services, education, and government deploying AI in contexts affecting individual decisions, this is a watch signal. Does your institution have rules governing the accuracy standards your AI tools must meet — and a process for evaluating when they do not?
The Ungoverned lesson: Policy governance means your institution has defined accuracy standards for AI outputs before a regulator defines them for you.
Source: FTC, July 2026 — ftc.gov/news-events/news/press-releases/2026/07/ftc-seeks-public-comment-policy-statement-addressing-ai-accuracy
Signal 8 — Protections: A Khosla-backed startup ran a 27-billion-parameter Alibaba model on an iPhone. Your device policy has a new governance question.
PrismML announced it has shrunk Qwen 3.6 — an open-source large language model developed by Chinese internet giant Alibaba — to run on an iPhone 17 Pro. The model has 27 billion parameters and is capable of complex chat, reasoning, fully autonomous agents, and software coding. When frontier-capable AI built on a Chinese-developed open-source model runs on an employee’s or student’s personal device, what are the red lines? What data can that agent access? Most institutional device policies were not written for a 27-billion-parameter autonomous agent running locally on a consumer device — let alone one whose underlying model may itself carry emergent capabilities still being discovered.
The Ungoverned lesson: Protections governance means your red lines are defined for AI capabilities that exist today — not the capabilities you anticipated when the policy was written.
Source: The Information, 2026 — theinformation.com/articles/khosla-backed-startup-claims-breakthrough-largest-ever-ai-model-iphone
Signal 9 — Privacy: The EU’s data protection body issued guidance on anonymization and web scraping for generative AI.
The European Data Protection Board issued guidance clarifying when data anonymization is sufficient for generative AI training and when web scraping for AI training constitutes unlawful personal data processing under GDPR. For institutions that use generative AI tools whose training data includes scraped web content — which applies to most major frontier models — this guidance has direct privacy governance implications. Does your institution know how the AI tools it deploys were trained, what data was used, and whether that training data may trigger GDPR or HIPAA obligations?
The Ungoverned lesson: Privacy governance means your institution has mapped the provenance of AI training data — not just AI output data flows.
Source: European Data Protection Board, July 2026 — edpb.europa.eu/news/edpb-sheds-light-on-anonymisation-and-web-scraping-for-generative-ai-and-adopts-final-version_en
Signal 10 — Provenance: Fewer than 1 in 3 organizations can demonstrate AI governance in practice. The evidence gap is now documented.
DigiCert’s AI Trust Pulse provides the first enterprise-scale evidence baseline on AI governance implementation: despite near-universal acknowledgment that AI governance is a priority, fewer than one in three organizations have implemented governance in practice. The Provenance dimension asks whether your institution can demonstrate, with documented evidence, that its AI tools meet the trustworthiness standards it claims to uphold. Most cannot. That is not a reputational risk. It is a fiduciary one — and when emergent capabilities are being discovered in those same tools, the fiduciary exposure compounds.
The Ungoverned lesson: Provenance governance means your institution can demonstrate — with evidence, not assertion — that its AI tools meet the standards it claims to uphold.
Source: DigiCert AI Trust Pulse, 2026 — digicert.com/content/dam/digicert/pdfs/report/ai-trust-pulse.pdf
Signal 11 — Preparedness: CHAI released the first agentic AI healthcare protocol guide. Clinical AI Governance as Leadership has a new standard.
The Coalition for Health AI published its Best Practice Guide for Agentic AI, which includes healthcare-specific protocol language considerations and provides the first sector-specific governance framework for AI agents operating in clinical environments. For health system boards, CMIOs, and clinical informatics leaders, this guide establishes the preparedness baseline: agent authorization in clinical contexts, delegation boundaries for clinical AI agents, and human checkpoint requirements when agents influence clinical decisions. Preparedness governance means your institution’s clinical AI governance has been assessed against this standard before an agent makes a consequential clinical decision without documented authorization.
The Ungoverned lesson: Preparedness in healthcare AI is no longer theoretical — CHAI has defined what clinical governance readiness looks like. The question is whether your institution has built it.
Source: Coalition for Health AI, 2026 — chai.org · assets.ctfassets.net/7s4afyr9pmov/7BBJg5qVgDll6MwONJIEX2/bdd239e89cb8b31569c7d90459041b28/Best_Practice_Guide_-Agentic_AI__v1.0.pdf
Signal 12 — Planet: The IMF updated its World Economic Outlook with AI at the center. Infrastructure and energy implications are now macroeconomic in nature.
The International Monetary Fund’s July 2026 World Economic Outlook update places AI investment and its infrastructure implications at the center of global economic projections. AI data center buildout — currently projected at approximately $650 billion annualized — is now a material factor in global energy demand, critical mineral supply chains, water consumption, and carbon emissions trajectories. Institutions with significant AI infrastructure dependencies are now exposed to energy cost volatility, supply chain disruption risk, and regulatory pressure on AI’s environmental footprint that did not exist three years ago.
The Ungoverned lesson: Planet governance means your institution has assessed the infrastructure and energy implications of AI as operational risks — not just environmental ones.
Source: IMF, July 8, 2026 — imf.org/en/publications/weo/issues/2026/07/08/world-economic-outlook-update-july-2026
The Common Thread
Every signal this week maps to the same pattern: governance is not keeping pace with deployment — and emergent capabilities in complex AI systems mean the gap is structural, not incidental.
Anthropic is publishing research about emergent properties in its deployed systems. The Treasury is warning about systemic risk privately while staying publicly bullish. Illinois is legislating to address the governance gap, while federal frameworks remain absent. DigiCert has measured the gap at the enterprise level. The UN Panel has documented it at global scale.
The Seba 12 Ps of Responsible AI Oversight — Purpose through Product Ownership — provide the institutional framework for closing it, one governance dimension at a time. And AI Governance as Leadership means that the institution that chooses to lead does not wait for the vendor to finish discovering what its system is before building the architecture to govern it.
The Board-Ready Action
The Seba UN Panel Governance Gap Assessment
Five questions drawn directly from the UN Scientific Panel’s preliminary report findings. One page. Built to be shared at your next leadership meeting.
Question 1 — Independent verification: The Panel found that the safety evaluation is largely designed by the companies being evaluated. What is your institution’s independent verification mechanism? If none exists, what is your governance floor?
Question 2 — Agentic AI oversight: The Panel found current oversight methodologies cannot adequately assess multi-agent systemic risks. Does your institution have agentic AI in deployment? What is the oversight methodology — and was it designed for the autonomy level the agents actually operate at?
Question 3 — Supply chain provenance: Has your institution assessed the provenance of its AI tools — including which systems were built on distilled or open-source models with non-domestic origins — and what that means for the governance assumptions you are relying on?
Question 4 — Performance equity: Has your institution assessed how its AI tools perform across the linguistic, cultural, and demographic populations it serves — not just the English-language benchmark populations on which most models were evaluated?
Question 5 — Governance sequencing: Has your institution built AI Governance as Leadership before deployment — or is it building governance in response to what has already been deployed?
If your leadership team cannot answer all five cleanly, your institution has a governance gap the Panel has now documented at global scale. Closing it is not a technology project. It is a leadership one.
Save this. Share it with your board.
What I Am Watching
Whether Anthropic’s global workspace research produces follow-on governance guidance — and whether other frontier labs publish similar research about emergent capabilities in their deployed systems, forcing a sector-wide reckoning with what institutional Product Ownership means in the context of emergence.
Whether the Treasury AI bubble report reaches its intended public audience — and whether financial regulators begin requiring AI infrastructure exposure disclosure from institutions with significant AI dependencies.
Whether Illinois’s mandatory annual audit requirement becomes the model for other states — and whether the California, New York, Illinois triad effectively forces a national compliance standard before federal legislation arrives.
Whether CHAI’s agentic AI healthcare protocol guide becomes the clinical governance baseline that health system accreditors reference in their 2027 standards.
On the desk for the Applied Frameworks miniseries continues this Wednesday with the EU AI Act — the world’s first comprehensive binding AI law. You do not need an office in Europe to be bound by Europe. Coming next: China, Singapore, South Korea, Canada, UK, Japan, UAE, and beyond. Full series at freddieseba.com.
Closing Thought
In complex systems science, emergence is not a surprise. It is a property of the system. If one emergent capability has been identified and documented in a deployed AI system, the governance assumption must be that others exist — at various stages of discovery, description, and disclosure.
The institutions that will govern AI well in this era are not the ones that understood every capability before deployment. They are the ones that built the governance architecture to own the consequences of what they did not yet know — and named the accountable human before the research paper was published.
That is AI Governance as Leadership. The Panel has documented the gap. Twelve signals this week show it widening. The work of closing it remains human.
Gratitude and Acknowledgments
This issue draws on the work of the Independent International Scientific Panel on Artificial Intelligence — 40 independent scientists operating under a non-political mandate, presenting to 193 UN member states. The Panel’s preliminary report is available in open access at un.org.
Appreciation this week to the Coalition for Health AI for the agentic AI healthcare protocol guide — clinical AI Governance as Leadership when it is most needed. To AMIA — the American Medical Informatics Association — whose foundational work on health data stewardship, clinical informatics ethics, and AI in clinical decision support continues to inform institutional governance practice in healthcare. To DigiCert for the AI Trust Pulse. To Capitol News Illinois for nonpartisan coverage of the Illinois AI Safety Measures Act. To NOTUS for surfacing the Treasury internal report. To the communities that shape this work week after week: AMIA, the University of San Francisco, the University of Illinois Chicago, and the boards, trustees, clinicians, executives, and faculty who put these frameworks to work in real institutions.
This issue also draws on Ungoverned: Applied Frameworks Under the Lens — an ongoing miniseries that applies the AI-MVG framework and the Seba 12 Ps of Responsible AI Oversight to organizations shaping AI governance globally — full series at freddieseba.com.
References
Anthropic. (2026). Global workspace research. https://www.anthropic.com/research/global-workspace
Axios. (July 6, 2026). Anthropic says Claude has internal thinking space, stopping short of calling it conscious. https://www.axios.com/2026/07/06/anthropic-claude-ai-conscious
Independent International Scientific Panel on Artificial Intelligence. (2026, July). Preliminary report: Evidence-based assessment of opportunities, risks and impacts of artificial intelligence. United Nations. https://www.un.org/digital-emerging-technologies
Katz, E. (July 6, 2026). Treasury has an internal report warning about the dangers of an AI bubble. NOTUS. https://www.notus.org/economy/treasury-internal-report-warning-dangers-ai-bubble
Dougherty, M. (July 6, 2026). Pritzker signs landmark AI regulation bill that aims to mitigate risks. Capitol News Illinois. https://capitolnewsillinois.com/news/pritzker-signs-landmark-ai-regulation-bill-that-aims-to-mitigate-risks
Reuters. (July 6, 2026). Microsoft joins AI-driven tech layoff wave with 4,800 job cuts. https://www.reuters.com/business/world-at-work/microsoft-joins-ai-driven-tech-layoff-wave-with-4800-job-cuts-2026-07-06
DigiCert. (2026). AI Trust Pulse. https://www.digicert.com/content/dam/digicert/pdfs/report/ai-trust-pulse.pdf
Federal Trade Commission. (2026, July). FTC seeks public comment on policy statement addressing AI accuracy. https://www.ftc.gov/news-events/news/press-releases/2026/07/ftc-seeks-public-comment-policy-statement-addressing-ai-accuracy
The Information. (2026). Khosla-backed startup claims breakthrough: largest ever AI model on iPhone. https://www.theinformation.com/articles/khosla-backed-startup-claims-breakthrough-largest-ever-ai-model-iphone
European Data Protection Board. (2026, July). EDPB sheds light on anonymization and web scraping for generative AI. https://www.edpb.europa.eu/news/edpb-sheds-light-on-anonymisation-and-web-scraping-for-generative-ai-and-adopts-final-version_en
Coalition for Health AI. (2026). Best Practice Guide — Agentic AI: Healthcare-specific protocol language considerations (v1.0). https://assets.ctfassets.net/7s4afyr9pmov/7BBJg5qVgDll6MwONJIEX2/bdd239e89cb8b31569c7d90459041b28/Best_Practice_Guide_-_Agentic_AI__v1.0_.pdf
International Monetary Fund. (July 8, 2026). World Economic Outlook Update — July 2026. https://www.imf.org/en/publications/weo/issues/2026/07/08/world-economic-outlook-update-july-2026
Seba, F. (2026). Ungoverned: A Practical Guide to AI Minimum Viable Governance. Amazon. https://www.amazon.com
About Dr. Freddie Seba
Dr. Freddie Seba helps boards, trustees, and executive leadership teams build practical AI governance before AI failures make governance unavoidable.
Scholar-operator, Silicon Valley founder, and global executive — EdD · USF · MBA · Yale · MA · Stanford — he translates fast-moving AI developments into governance frameworks leaders can deploy now, across healthcare, financial services, and higher education. Non-vendor. Non-partisan. Doctoral research, not advocacy.
Ungoverned: A Practical Guide to AI Minimum Viable Governance is available now on Amazon.
Booking keynotes and workshops for fall 2026 — freddieseba.com/contact
Transparency
Drafted and refined with AI-assisted tools — this issue, Anthropic’s Claude Sonnet 4.6. Naming the tools we use is part of the provenance we advocate. This is disclosure, not endorsement. Final editorial control and responsibility remain with the author.
For informational and educational purposes only. It does not constitute legal, regulatory, or compliance advice.
For reprint or licensing inquiries: contact@freddieseba.com
© 2026 Freddie Seba. All rights reserved.
Connect
freddieseba.com · LinkedIn: @freddiesebaprofile · YouTube · Spotify · Apple Podcasts: AI Governance with Dr. Freddie Seba
#AIGovernance #Ungoverned #AIMVG #Seba12Ps #BoardGovernance #AgenticAI #AILeadership #ResponsibleAI #UNAIPanel #HealthAI #CIO #GeneralCounsel
anthropic.com/research/global-workspace
Leave a Reply