TL;DR
Andreessen Horowitz published a proposal on November 12 for a CEO-led effort to protect selected operators and shared vendors in five U.S. infrastructure sectors from AI-enabled cyber threats. The plan sets proposed security targets and reporting duties, but it is not policy; costs, participation and government adoption remain unresolved.
Venture firm Andreessen Horowitz (a16z) published a proposal on November 12 calling for CEOs in five critical infrastructure sectors to commit to cybersecurity targets, as AI could shorten the time between finding a vulnerability and exploiting it. The plan would put a White House task force in charge of setting targets, with industry groups reporting progress; it remains a proposal, not government policy.
Get the latest gadgets delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
The plan, titled “Super Intelligence-Era Critical Infrastructure Cyber Defense Plan,” was written by Anne Neuberger, a16z general partner and head of global affairs. According to a16z, Neuberger also outlined it in a Wall Street Journal opinion piece. The proposal focuses on power, banking and payments, water, communications and hospitals, arguing that a limited number of operators and shared vendors could have outsized effects if compromised.
For each sector, the plan proposes three CEO-signed targets: zero internet-exposed control systems, successful 72-hour drills in manual operation, and containment of critical vulnerabilities within a set time. It says the highest-priority targets should be contained within six months. The proposal also calls for a joint cross-sector exercise within three to six months and monthly, anonymized scorecards from industry information-sharing centers, or ISACs.
Under the suggested governance structure, the White House’s Super Intelligence Force would set targets, a council of two or three CEOs per sector would answer for results, and ISACs would handle implementation reporting. A sector falling behind could face a CEO-level review; the council would meet the president every six months. The plan asks Congress to reauthorize the Cybersecurity Information Sharing Act of 2015 over the long term and considers a potential safe harbor for companies that meet outcomes-based targets.
The proposal would shift attention from trying to secure every system to prioritizing operators and vendors whose disruption could spread across essential services. Its focus on measurable outcomes could make progress easier to track across industries that currently have different systems and security practices.
The plan’s urgency rests on a claim about the changing pace of cyber risk: a16z argues that AI can help attackers find weaknesses faster, while defenders could use the same tools to identify exposed systems, rank risks and generate fixes. That is the firm’s rationale, not evidence that AI has already produced a specific incident or that the proposed approach would prevent one.
Any scheme that ties company commitments to federal oversight and possible legal protection could affect how infrastructure operators share threat information and disclose vulnerabilities. Whether those incentives would be sufficient, and whether the plan could be implemented across sectors with different ownership and regulatory arrangements, is not established in the proposal.
Products Worth Considering
Samsung 32-inch Odyssey G5 G51F QHD 180Hz Gaming Monitor
Bose QuietComfort Wireless Bluetooth Headphones – Active Over‑Ear Noise Cancelling with Mic, USB‑C Charging, Deep Bass, up to 24 h Playtime,
Logitech MX Keys S Combo Wireless Keyboard and Mouse, Graphite
As an affiliate, we earn on qualifying purchases.
A Proposal for the White House Task Force
The plan is addressed to the White House’s Super Intelligence Force. The source article says President Donald Trump announced the task force on October 4, 2026, and named National Intelligence Director Jay Clayton to lead it; NBC News was cited for that account. The proposal describes the task force as the federal body that would coordinate targets and receive industry scorecards.
Neuberger’s prior government roles are relevant to the proposal’s focus: she says she worked on infrastructure security at the NSA and later as deputy national security adviser for cyber and emerging technology. She is now at a16z, a venture firm that invests in frontier AI labs and security companies. That connection does not establish a conflict, but it is context for readers assessing the proposal and its recommendation to include companies in the process.
The plan says replacement of unsupported equipment could take two to five years, while it sets a six-month goal for containing the highest-priority targets. Those are timelines stated in the proposal, not a record of completed work or a confirmed government schedule.
“AI is shrinking the time between when a weakness is found and when it is exploited. Defenders who still work at human speed will lose that race.”
— Anne Neuberger, a16z general partner and head of global affairs, in the plan
Policy, Costs and Participation Unresolved
No government adoption, implementation schedule or participating companies are identified in the source material. The proposal gives no cost estimate, and it does not specify the precise set of operators that would qualify as the highest-priority targets or how target compliance would be independently verified.
The proposed legal safe harbor is described as a possibility, not an agreed protection. It is also unclear how the suggested targets would account for sector-specific rules, legacy equipment or operators that cannot meet the proposed deadlines. The plan’s claims about AI’s effects on both attacks and defenses are arguments made by a16z; the material does not quantify those effects or provide independent evidence that the proposed measures would work at the stated pace.
Whether Officials Adopt the Plan
The next milestone is whether the White House’s Super Intelligence Force or other federal officials respond to the proposal, and whether infrastructure companies or ISACs agree to participate. If adopted, the plan anticipates monthly sector scorecards, a cross-sector drill within three to six months and a six-month containment goal for top targets. Those milestones are proposed steps; no start date or official commitment is confirmed.
Key Questions
What did a16z propose?
A CEO-led cybersecurity effort for five sectors—power, banking and payments, water, communications and hospitals—with federal target-setting and industry progress reporting.
Is the plan already U.S. policy?
No. It was published by a16z on November 12, 2026, and is described as a proposal. The source material does not report that the government adopted it.
What targets would CEOs sign?
The proposed targets are zero internet-exposed control systems, passing 72-hour manual-operation drills and containing critical vulnerabilities within a set period. The plan says top targets should be contained within six months.
Who would oversee the effort?
The plan assigns target-setting to the White House’s Super Intelligence Force, accountability to a CEO council, and monthly reporting to sector ISACs.
What remains uncertain?
Government adoption, company participation, costs, target verification and the terms of any legal safe harbor are not settled in the proposal.
Source: Thorsten Meyer AI
