Rishi Sunak’s move into senior advisory roles at Microsoft and Anthropic marks one of the most consequential post-premiership pivots in recent UK political history — a high-profile marriage of political capital and cutting-edge AI ambition that raises urgent questions about the revolving door, corporate influence in public policy, and the governance of artificial intelligence at both national and global levels.
Rishi Sunak, who led the UK government from 2022 until his resignation as Conservative leader in 2024 and now remains a Member of Parliament, has accepted part‑time, paid senior advisory roles with Microsoft Corporation and Anthropic PBC. Both appointments were reviewed by the UK’s Advisory Committee on Business Appointments (Acoba), which placed conditions intended to reduce the risk of improper influence — most notably a ban on lobbying UK ministers and officials for two years and strict admonitions to avoid using privileged government knowledge. Sunak has said he will donate his remuneration from these roles to The Richmond Project, a charity he co‑founded.
These moves come against a backdrop of Sunak’s active role in shaping UK tech and AI policy while in office. He hosted the 2023 AI Safety Summit, established institutional infrastructure around AI safety, and presided over major deals that attracted large-scale cloud and compute investment into the UK — including Microsoft’s announced £2.5 billion AI data‑centre and skills investment unveiled in November 2023. That previous relationship between Sunak and Microsoft has been explicitly noted by regulators and commentators as part of the reason the appointments warranted scrutiny.
This episode should prompt a sober reappraisal of post‑office appointment rules, independent compliance mechanisms, and corporate practices that ensure advisers contribute strategic value without blurring the line between private advantage and public responsibility. The strength of AI governance — in the UK and globally — will increasingly turn on the quality of those institutional guardrails, not just the pedigree of the people who move between public office and private boardrooms.
Source: TechCrunch Former UK Prime Minister Rishi Sunak to advise Microsoft and Anthropic | TechCrunch
Rishi Sunak, who led the UK government from 2022 until his resignation as Conservative leader in 2024 and now remains a Member of Parliament, has accepted part‑time, paid senior advisory roles with Microsoft Corporation and Anthropic PBC. Both appointments were reviewed by the UK’s Advisory Committee on Business Appointments (Acoba), which placed conditions intended to reduce the risk of improper influence — most notably a ban on lobbying UK ministers and officials for two years and strict admonitions to avoid using privileged government knowledge. Sunak has said he will donate his remuneration from these roles to The Richmond Project, a charity he co‑founded. These moves come against a backdrop of Sunak’s active role in shaping UK tech and AI policy while in office. He hosted the 2023 AI Safety Summit, established institutional infrastructure around AI safety, and presided over major deals that attracted large-scale cloud and compute investment into the UK — including Microsoft’s announced £2.5 billion AI data‑centre and skills investment unveiled in November 2023. That previous relationship between Sunak and Microsoft has been explicitly noted by regulators and commentators as part of the reason the appointments warranted scrutiny.
What the appointments are — and what Acoba required
Roles and remit
- At Microsoft, Sunak’s role is described as internally focused, providing “high‑level strategic perspectives on macro‑economic and geopolitical trends and how they intersect with innovation, regulation, and digital transformation.” He will also speak at company events such as the Microsoft Summit. Microsoft and Sunak told Acoba the role will not involve advising on UK policy and that he will not initiate contact with UK government departments.
- At Anthropic, Sunak will advise on global strategy, macroeconomic, and geopolitical trends, again with the company and Acoba stressing the role’s internal orientation and restrictions on UK government engagement. Anthropic has framed the hire as part of a broader push to scale international operations and bring experienced strategic perspectives into its leadership.
Key Acoba conditions (summary)
- A prohibition on lobbying UK ministers, civil servants, or departments for two years.
- A requirement not to use privileged information acquired while in public office, and to avoid advising specifically on UK policy or UK government contracts.
- Agreements from both companies to ring‑fence the role and limit policy‑facing duties and contacts.
- Sunak’s pledge to donate his salary from both appointments to The Richmond Project, removing personal financial gain from immediate scrutiny.
Why this matters: immediate facts, verified
- Sunak’s appointments to Microsoft and Anthropic were publicly confirmed in October 2025 and reviewed by Acoba. Both companies and Sunak have publicly stated the roles are part‑time and internally focused.
- The watchdog’s decision explicitly referenced concerns that Sunak’s appointment could be perceived as granting unfair access or influence, given his recent premiership and the centrality of AI regulation debates. Acoba’s formal advice letters articulate those concerns and set conditions intended to mitigate them.
- Sunak has prior, demonstrable ties to Microsoft while in office, including his role in facilitating Microsoft’s November 2023 announcement of a £2.5 billion UK AI investment (AI datacentres, GPUs, and training initiatives). That deal remains a touchstone in discussions about whether former ministers accepting tech roles creates undue advantage.
Analysis: strategic value to Microsoft and Anthropic
What Sunak brings to the table
- Government experience at the highest level. Sunak’s understanding of geopolitical risk, international economic policy, and the inner workings of Whitehall is a rare asset for firms navigating regulatory headwinds and national security scrutiny. That perspective is attractive to global tech firms that must align product strategy with political realities.
- AI safety and regulatory credibility. Sunak helped raise the international profile of the UK on AI safety during his premiership. For a firm like Anthropic, which brands itself around safety and responsible deployment, an adviser with credibility on safety governance and diplomatic channels is strategically valuable.
- Access to elite networks. Beyond formal privileges, former leaders carry relationships across business, finance, and international politics that can facilitate introductions, fundraising, and client relationships — intangible benefits that companies prize when scaling globally.
Why both Microsoft and Anthropic might want the same adviser
- Complementary advantages: Microsoft offers cloud, enterprise distribution, and deep market reach; Anthropic offers frontier AI research and competitive model development. An adviser who understands both corporate strategy and geopolitics can help each firm refine their international positioning, navigate public affairs, and anticipate regulatory trajectories.
- Perception vs. operational overlap: Both firms stress Sunak will not advise on UK policy or contracts. Operationally, an adviser focused on macro trends is not the same as a lobbyist. But in practice, strategic framing and event representation can indirectly shape policy conversations and ecosystem perceptions.
Conflict, perception, and the revolving door problem
Why the optics matter
Even if all contractual prohibitions are respected, the presence of a former prime minister on the payroll of major AI actors invites legitimate concern about perceived access. Perception is not a trivial reputational externality: it influences public trust, parliamentary scrutiny, and the willingness of other governments to engage. Acoba itself flagged that perception as a central concern in its guidance.Structural risks
- Information asymmetry: Former prime ministers inevitably retain contextual knowledge about priorities, risk assessments, and the internal calculus of government decision‑making. While regulators can bar the use of explicit privileged information, unwritten knowledge — relationships, timing expectations, policy instincts — is harder to fence off and can influence corporate strategy in subtle ways.
- Regulatory capture by proximity: The longer former senior officials gravitate towards corporate boards or advisory rosters in the same sectors they once regulated, the greater the risk that industry actors shape the policy agenda through second‑hand channels. This creates a feedback loop where regulators adapt to industry norms rather than asserting independent public‑interest policy.
- Competing client conflicts: Advising two firms that are competitors or occupy adjacent positions in the AI stack raises governance questions about confidentiality, conflict management, and the adviser’s ability to remain truly neutral. Microsoft’s deep commercial ties to OpenAI and Anthropic’s position as a competitor intensify that issue.
Enforcement limits
Acoba’s restrictions (two‑year lobbying bans, no use of privileged information) depend on voluntary compliance plus post‑hoc public accountability. Historically, critics of Acoba argue the committee lacks teeth — it can recommend conditions, but enforcement and sanctions are limited. That structural limitation matters here, because the reputational and informational advantages of hiring a former head of government are not easily reversible.Benefits and legitimate arguments for the hires
While concerns are real, there are defensible, concrete reasons why both companies might consider these hires appropriate and even justified:- Legitimate need for policy‑fluent strategic advice. Global tech firms operate in a regulatory labyrinth that includes trade controls, export licensing, national security reviews, and data‑localisation mandates. Former senior public servants can provide legitimate macroeconomic and geopolitical insight that helps firms align investments with policy realities without stepping across legal boundaries.
- AI safety expertise. If advisers focus on stewardship, safety frameworks, and public‑interest alignment rather than lobbying, they can accelerate the development of governance best practices inside industry, potentially complementing public policy rather than undermining it. Anthropic’s safety framing and Sunak’s prior engagement on AI safety create a plausible intersection where both parties can contribute to public goods — if handled transparently.
- Philanthropic offset and transparency. Sunak’s pledge to donate remuneration to The Richmond Project reduces personal financial incentive. Public disclosure of the Acoba advice letters and corporate confirmations adds a layer of transparency that would not exist in a private appointment. That transparency is a mitigant, albeit not a cure.
Practical governance checklist — what meaningful guardrails should look like
To move from words to enforceable practice, firms, former ministers, and regulators should adopt a concrete checklist for high‑risk appointments:- Publicly publish the adviser’s scope, limitations, and meeting logs where they interact with public‑sector representatives.
- Institute firewalling inside the company: separate teams and formal reporting lines that prevent access to commercial negotiation teams or bid desks that handle government contracts.
- Independent oversight: create a third‑party compliance attestation every six months confirming the adviser has not engaged in prohibited lobbying or used privileged information.
- Cooling‑off periods longer than two years for particularly sensitive sectors (AI safety and national security infrastructure).
- Mandatory recusal for the adviser from any work that could plausibly intersect with decisions made under their term in office.
Geopolitics, competition, and the global AI race
Strategic context
The appointments occur during a heated phase in the global AI competition. Nations are designing industrial strategies around compute, data sovereignty, and AI safety frameworks. Corporates are simultaneously racing to own model capabilities, enterprise distribution channels, and international regulatory goodwill. Hiring high‑level political advisers is a rational response to that environment: a way to anticipate policy shifts, secure market access, and interpret geopolitical flashpoints.A high‑stakes balancing act
- For governments, the balance is delicate: welcoming private investment and expertise while safeguarding policy independence and public interest.
- For companies, long‑term legitimacy depends not only on product performance but on being perceived as trustworthy partners in public safety, not rent‑seekers. That calculus determines competitive advantage as much as technical prowess.
Precedents and the broader pattern
This is not an isolated case. The past decade has seen senior politicians take roles in Big Tech and finance after leaving office, including high‑visibility moves such as:- Former UK deputy prime minister Nick Clegg moving to Meta as president of global affairs (and stepping down in January 2025), a pattern of political figures joining tech firms to manage global policy and reputational risks.
- Sunak’s own wider post‑office engagements, including advisory roles at Goldman Sachs and paid speeches for private firms, reflect a broader trend of former officials monetizing experience in advisory capacities.
- The appointment of Sunak’s former chief of staff, Liam Booth‑Smith, to Anthropic earlier in 2025 underscores the interweaving of political networks with the AI industry and the recurring need for Acoba to adjudicate potential conflicts.
Risks to watch — short term and long term
Short term
- Perception‑driven political fallout. Immediate parliamentary questions, media scrutiny, and opposition pressure can force additional constraints and distract governments at critical moments in AI regulation.
- Contractual conflicts. Even if Sunak is barred from lobbying for two years, secondary interactions (conferences, roundtables, private introductions) can create soft influence channels that undermine the spirit of those restrictions.
Long term
- Erosion of public trust. If the public senses a pattern where former leaders systematically join firms they once regulated, confidence in impartial governance may erode, complicating future policymaking in technology and beyond.
- Policy capture risk. Over time, a persistent channel of former officials into industry creates institutional memory and norms that can bias regulatory design toward industry preferences, making robust safety standards harder to implement.
- International precedent. Other countries may emulate weaker safeguards if high‑profile appointments become normalized, undermining coordinated international approaches to AI safety and trade controls.
What to expect next
- Parliamentary and public scrutiny. Expect further questions in the UK Parliament, additional media reporting, and NGO commentary assessing whether the Acoba conditions are adequate and enforceable.
- Operational firewalls. Both Microsoft and Anthropic will likely document and publicize the compliance measures they have implemented to show they are meeting Acoba’s conditions.
- Possible policy responses. The episode feeds into broader debates about strengthening the UK’s post‑employment rules for senior officials — including proposals for longer cooling‑off periods, greater transparency, and stronger enforcement mechanisms.
Conclusion
Rishi Sunak’s advisory appointments to Microsoft and Anthropic crystallize the tensions at the heart of modern tech governance: the need for governments and companies to exchange expertise, and the risk that such exchanges create real or perceived channels of influence that can erode public trust. The Acoba conditions and Sunak’s philanthropic pledge are important mitigants, but they do not eliminate the structural question: how democracies preserve the integrity of policy‑making while tapping into the knowledge of those who once held its levers.This episode should prompt a sober reappraisal of post‑office appointment rules, independent compliance mechanisms, and corporate practices that ensure advisers contribute strategic value without blurring the line between private advantage and public responsibility. The strength of AI governance — in the UK and globally — will increasingly turn on the quality of those institutional guardrails, not just the pedigree of the people who move between public office and private boardrooms.
Source: TechCrunch Former UK Prime Minister Rishi Sunak to advise Microsoft and Anthropic | TechCrunch