← All posts / Policy

Five Mission Centers and a 30-Day Clock: Inside the NSA's Biggest Restructuring in a Decade

NSA director Gen. Joshua Rudd is recasting the world's largest electronic spy agency into five mission centers — AI, China, cybersecurity, warfighting support and global intelligence — on a 30-day implementation clock, with full operational capability targeted for January 2027.

Five Mission Centers and a 30-Day Clock: Inside the NSA's Biggest Restructuring in a Decade

The National Security Agency is undertaking its most extensive internal restructuring in at least a decade, and artificial intelligence is getting its own seat at the table. According to a Washington Post exclusive published September 13 and confirmed by Recorded Future News, NSA director Army Gen. Joshua M. Rudd — who also heads U.S. Cyber Command — has begun recasting the world’s largest electronic spy agency into five new “mission centers,” devoted to China, cybersecurity, artificial intelligence, combat support, and global intelligence.

The news landed on a weekend when AI and national security were already colliding across headlines: China’s state security minister publicly framing AI as a “new arena for strategic rivalry among major powers,” and the Senate’s frontier-AI duty-of-care debate still echoing through Washington. Against that backdrop, the NSA putting AI alongside China and cybersecurity as a standalone mission — rather than a supporting technology buried inside a directorate — is one of the clearest organizational signals yet of how seriously the U.S. intelligence community now takes machine intelligence, both as a tool and as a threat surface.

What is actually changing

The restructuring replaces the NSA’s existing directorates with five mission centers headquartered at Fort Meade, Maryland. Each will be led by a newly elevated “mission director” who, according to a former NSA official cited by Chosun Biz, will carry de facto authority on par with the NSA deputy director. That is a significant concentration of power: these are not coordination bodies but operational commands.

The five missions:

  • Artificial intelligence — a dedicated center for AI, the first time the agency has given the technology its own mission-level organization
  • China — a single home for what has long been the agency’s highest-priority target
  • Cybersecurity — consolidating the NSA’s defensive mission, including its work securing U.S. national security systems
  • Combat support / warfighting — the pipeline that gets NSA intelligence to military commanders and real-world battlefields faster
  • Global intelligence — the rest of the world, and notably the new home for Tailored Access Operations

Some major details remain unresolved. It is unclear whether the five centers will replace existing divisions outright or layer on top of the current structure. The fate of newer organizations like the Cybersecurity Collaboration Center, which handles threat-intelligence sharing with private industry, is hazy. And the reorganization’s effect on the NSA’s relationship with U.S. Cyber Command — the two share resources and a campus at Fort Meade under Rudd’s dual-hat leadership — is still being worked out.

The speed is the story

Rudd shared the changes with the NSA workforce earlier this month, starting what sources describe as a 30-day clock to implement them. The new mission directors must submit their restructuring plans to Rudd by the end of September. Public announcement is expected in mid-October, with the five centers reaching “full operational capability” — the military term meaning each can fulfill its required missions — by January 2027.

For an agency of more than 30,000 military and civilian employees famous for its slow-moving bureaucracy, the pace is aggressive almost to the point of recklessness. Both a current and a former agency official told Recorded Future that Rudd and NSA Deputy Director Tim Kosiba know the rapid realignment will “break things” internally and plan to adjust on the fly.

That tone comes straight from the top. In his second confirmation hearing before the Senate Intelligence Committee in January, Rudd — a decades-long Special Forces veteran eventually confirmed to the dual NSA/Cyber Command role in March — said his “first priority is speed. Second one is scale, the next one is innovation and then integration. We’ve got to move as fast as we can to ensure that we’ve got the right technologies, the right capabilities and [be] able to generate multiple options.”

TAO’s return and a bigger budget

One of the most notable specifics: Tailored Access Operations, the NSA’s elite hacking unit, will be placed under the global intelligence center. TAO was originally created in 1997 to compromise the hardest-to-reach digital targets on Earth. Its name was retired in 2016 and its members dispersed across the agency — until Rudd reconstituted the unit in recent months. According to the Washington Post’s reporting, TAO is expected to receive a significantly increased budget when the new fiscal year begins on October 1.

Resurrecting a storied offensive unit and then immediately giving it a bigger budget and a new home inside the org chart says a great deal about the agency’s priorities: offensive access capabilities remain central, and the global intelligence mission center will be one of the most powerful fiefdoms in the new structure.

The ghosts of NSA21 and EVO25

This is not the agency’s first attempt at reinvention, and the history explains the internal anxiety. The last restructuring of comparable scope, roughly a decade ago, was “NSA21,” which sought to meld the agency’s offensive and defensive organizations. Years later, it is remembered inside Fort Meade as a failure that generated more bureaucratic confusion than capability — and the current reshuffle is already triggering frustration and anger among current and former officials who lived through it.

A more recent effort, “EVO25,” began around the start of the second Trump administration and would have consolidated the agency’s offices. Many read it as a defensive move to avoid having a restructuring imposed from outside by the U.S. DOGE Service; it was ultimately abandoned. Rudd’s plan is explicitly a break from both: rather than consolidating for consolidation’s sake, it carves the agency into mission-aligned commands and accepts breakage as the cost of speed.

There is also a personnel signal embedded in the plan. Rudd told senior officials that the heads of the five new mission centers might be selected from outside the agency — a prospect one LinkedIn summary of the Post’s reporting described as “unnerving NSA” veterans. Some picks, including the new head of AI, have already been made and could be announced internally as soon as this week.

Why an AI mission center matters

It is tempting to read “AI center” as a research arm. The context suggests something broader. The NSA’s core product is signals intelligence — collecting and making sense of planetary-scale communications traffic — and that is precisely the workload where modern AI models excel and where AI-powered adversaries now operate. The same weekend the restructuring broke, China’s intelligence chief was warning publicly about AI-driven threats to state data and business secrets, and Western security agencies have spent the year documenting AI-accelerated intrusion campaigns, from agent-driven PaperCut exploitation to the distillation attacks on frontier model providers that prompted a joint NSA/CISA/FBI advisory.

A dedicated AI mission center gives the NSA a single accountable organization for three converging jobs: harnessing AI to triage and exploit the intelligence it collects, defending U.S. systems against AI-enabled adversaries, and keeping pace with how targets — state and non-state alike — are industrializing their own use of machine intelligence.

Analysis: an agency betting it can move at AI speed

The deeper story here is institutional velocity. The intelligence community’s standard criticism of itself is that it reorganizes constantly while adversaries simply adapt. Rudd’s plan is a wager that this time structure can buy speed: mission directors with deputy-director-level authority, a 30-day clock, plans due within weeks, and full capability in under four months. If it works, the NSA gets meaningfully faster at turning collection into military effect at the exact moment AI is compressing decision timelines. If it fails, it becomes NSA21’s sequel — another decade remembered for churn.

Either way, the symbolism is unambiguous. When the world’s largest electronic spy agency formally elevates artificial intelligence to the same organizational tier as China and cybersecurity, it is declaring that AI is no longer a technology the agency uses. It is a domain the agency must win.