Washington's new private offensive cyber program offers vetted security vendors an untested shield against criminal prosecution while leaving their clients and non-participating companies exposed to substantial legal, insurance, and commercial risk, according to an analysis published on CSO Online. The August 12 National Security Presidential Memorandum directs the National Coordination Center to build a program allowing approved "Participating Companies" to conduct covert surveillance operations or carry out manipulation, disruption, degradation, or destruction of systems after two federal co-executive directors approve each operation in writing. The White House frames the initiative as consumer protection, citing more than $20.8 billion in American losses to cyber-enabled crime in 2025, but the memorandum functions as a risk-transfer contract where most parties bearing the risk never signed anything.
The program's criminal protection relies on a single untested interpretation of the Computer Fraud and Abuse Act, which exempts "lawfully authorized investigative, protective, or intelligence activity of a law enforcement agency" at 18 U.S.C. 1030(f). The memorandum styles every operation as federal law enforcement activity to fit inside that exemption, though Congress wrote that language for law enforcement agencies themselves and no court has ruled whether it extends to private companies operating under contract. The memorandum withholds far more protection than it provides: no civil safe harbor against lawsuits from collateral victims, no preemption of state anti-hacking laws, no protection under foreign law, and no indemnification. The Justice Department and Department of Homeland Security can require participating companies to post a forfeitable bond of at least $1 million as a contract condition, and the operating procedures governing day-to-day execution remain unpublished until mid-October.
The analysis finds that non-participating companies inherit exposure through four channels. Criminal groups rent and compromise the same cloud infrastructure, content delivery networks, and SaaS platforms that ordinary workloads occupy, so an approved effects operation against that substrate can surface as an unexplained outage in unrelated environments. Participating companies must disclose their commercial agreements to the National Coordination Center but face no customer-disclosure obligation, forcing buyers to extract written representations themselves. Lloyd's market bulletin Y5381 requires syndicates to carry state-backed cyberattack exclusions in standalone cyber policies, and standard clauses key attribution to government determinations, placing retaliation events or collateral losses from government-directed operations directly into exclusion analysis. The memorandum invites participating companies to buy threat information from private entities and propose operations built on it, meaning threat intelligence shared with ISACs, government channels, or commercial platforms can feed offensive proposals wherever receiving-party contracts permit that use.
The sharpest market evidence predates the memorandum itself. Reuters reported in January 2026 that Beijing directed Chinese firms to stop using cybersecurity software from roughly 15 American and Israeli vendors, and in February that Palo Alto Networks softened its attribution of a Chinese espionage campaign over concerns about retaliation against its personnel or clients. On August 6, six days before the signing, the Cyberspace Administration of China opened a formal cybersecurity review of Palo Alto's products using the same mechanism that previously barred Micron from Chinese critical-infrastructure procurement. The analysis warns that the same government-control language supporting the domestic criminal defense also strengthens the case for attributing operations to the United States internationally under Article 8 of the International Law Commission's state-responsibility articles, which treats private conduct as state action when it follows government instructions or runs under its direction and control.
The report directs corporate boards to answer five questions before the October operating rules arrive: which critical security, cloud, identity, and incident-response vendors intend to participate and will represent that status in writing; whether each participating vendor can segregate customer data from operations work and contract to that segregation; which representations to customers, regulators, and insurers become incomplete if a vendor participates without disclosure; what cyber policies pay on retaliatory state-backed attacks, accidental American-directed effects, and shared-cloud outages; and which employees, affiliates, and joint ventures in China or rival jurisdictions connect to participating vendors and what their travel protocol requires. The analysis recommends watching the October operating procedures and the Cyber Letters of Marque and Reprisal Act, introduced July 15 as S. 5000 and H.R. 9697, whose House version would supply a statutory civil shield the memorandum doesn't contain and a presidential instrument can't create. Participation remains each vendor's decision, but the government authorizes the operation while residual legal, insurance, and commercial exposure stays private, including with companies that never signed anything. The clearest immediate vulnerability lies in the silence: absent a standard attestation framework, vendor nationality itself becomes a rational procurement screen for foreign buyers unwilling to inherit geopolitical risk through the supply chain.