U.S. Diplomacy
Since its founding in 1922, Foreign Affairs has actually been the leading forum for serious conversation of American diplomacy and international affairs. The publication has actually featured contributions from numerous leading international affairs experts.

More Resources

- Feedback
- Institutional Subscriptions
- Gift a Subscription
- About Us
- Events
- Issue Archive
- Advertise
- Audio Content
- Account Management
- FAQs
Spy vs. AI

ANNE NEUBERGER is Deputy Assistant to the President and Deputy National Security Adviser for Cyber and Emerging Technology on the U.S. National Security Council. From 2009 to 2021, she served in senior functional functions in intelligence and cybersecurity at the National Security Agency, including as its first Chief Risk Officer.

- More by Anne Neuberger
Spy vs. AI

How Artificial Intelligence Will Remake Espionage

Anne Neuberger

-.
Copy Link Copied. Article link: https://www.foreignaffairs.com/united-states/spy-vs-aihttps://www.foreignaffairs.com/united-states/spy-vs-[ai](http://fabiennearch-psy.fr). Copy


Gift Link Copied. This is a subscriber-only function. Subscribe now or Sign in.


Create Citation Copied. Chicago MLA APSA APA. Chicago Cite not available at the moment. MLA Cite not available at the minute. APSA Cite not available at the moment. APA Cite not available at the minute


Download PDF. This is a subscriber-only function. Subscribe now or Check in.


Request Reprint. Request reprint permissions here.




In the early 1950s, the United States faced a vital intelligence challenge in its burgeoning competitors with the Soviet Union. Outdated German reconnaissance photos from World War II might no longer provide adequate intelligence about Soviet military capabilities, and existing U.S. monitoring abilities were no longer able to penetrate the Soviet Union's closed airspace. This deficiency spurred an audacious moonshot initiative: the advancement of the U-2 reconnaissance aircraft. In only a few years, U-2 objectives were delivering important intelligence, capturing pictures of Soviet missile setups in Cuba and bringing near-real-time insights from behind the Iron Curtain to the Oval Office.

Today, the United States stands at a comparable juncture. Competition in between Washington and its rivals over the future of the global order is magnifying, and now, much as in the early 1950s, the United States must take benefit of its world-class personal sector and sufficient capability for development to outcompete its foes. The U.S. intelligence neighborhood need to harness the country's sources of strength to deliver insights to policymakers at the speed of today's world. The integration of expert system, especially through large language designs, uses groundbreaking opportunities to enhance intelligence operations and analysis, allowing the delivery of faster and more pertinent assistance to decisionmakers. This technological transformation features significant drawbacks, however, specifically as enemies exploit comparable developments to discover and counter U.S. intelligence operations. With an AI race underway, the United States should challenge itself to be first-first to gain from AI, first to protect itself from opponents who might use the technology for ill, and first to use AI in line with the laws and worths of a democracy.

For the U.S. nationwide security neighborhood, satisfying the guarantee and managing the peril of AI will need deep technological and cultural modifications and a determination to change the way companies work. The U.S. intelligence and military communities can harness the capacity of AI while mitigating its fundamental dangers, guaranteeing that the United States maintains its competitive edge in a rapidly developing worldwide landscape. Even as it does so, the United States need to transparently communicate to the American public, and to populations and partners all over the world, how the nation means to fairly and safely use AI, in compliance with its laws and values.

MORE, BETTER, FASTER

AI's capacity to revolutionize the intelligence community depends on its capability to process and examine large quantities of data at unmatched speeds. It can be challenging to evaluate large quantities of gathered information to produce time-sensitive cautions. U.S. intelligence services might leverage AI systems' pattern recognition abilities to determine and alert human experts to potential dangers, such as missile launches or military motions, or crucial global developments that analysts understand senior U.S. decisionmakers are interested in. This capability would ensure that critical cautions are timely, actionable, and appropriate, permitting more reliable actions to both rapidly emerging dangers and emerging policy chances. Multimodal designs, which incorporate text, images, and audio, enhance this analysis. For circumstances, utilizing AI to cross-reference satellite imagery with signals intelligence might provide a detailed view of military movements, allowing faster and more accurate threat evaluations and potentially brand-new methods of providing details to policymakers.

Intelligence experts can also offload recurring and time-consuming jobs to devices to focus on the most fulfilling work: generating initial and deeper analysis, increasing the intelligence community's general insights and efficiency. A fine example of this is foreign language translation. U.S. intelligence agencies invested early in AI-powered abilities, and the bet has paid off. The capabilities of language designs have grown significantly sophisticated and accurate-OpenAI's just recently released o1 and o3 models showed substantial development in precision and reasoning ability-and can be utilized to a lot more quickly translate and summarize text, audio, and video files.

Although challenges remain, future systems trained on higher amounts of non-English information might be capable of critical subtle distinctions in between dialects and comprehending the meaning and cultural context of slang or Internet memes. By counting on these tools, the intelligence community might focus on training a cadre of extremely specialized linguists, who can be difficult to find, often battle to get through the clearance process, and take a very long time to train. And of course, by making more foreign language products available throughout the ideal agencies, U.S. intelligence services would be able to quicker triage the mountain of foreign intelligence they get to choose the needles in the haystack that truly matter.

The worth of such speed to policymakers can not be underestimated. Models can swiftly sift through intelligence data sets, open-source details, and standard human intelligence and produce draft summaries or preliminary analytical reports that experts can then validate and fine-tune, making sure the last products are both detailed and precise. Analysts might partner with an advanced AI assistant to overcome analytical problems, test concepts, and brainstorm in a collective fashion, enhancing each version of their analyses and historydb.date delivering finished intelligence quicker.

Consider Israel's experience in January 2018, when its intelligence service, the Mossad, discreetly got into a secret Iranian facility and took about 20 percent of the archives that detailed Iran's nuclear activities in between 1999 and 2003. According to Israeli authorities, the Mossad collected some 55,000 pages of files and a further 55,000 files stored on CDs, consisting of photos and videos-nearly all in Farsi. Once the archive was obtained, senior authorities put tremendous pressure on intelligence experts to produce detailed evaluations of its content and whether it pointed to a continuous effort to construct an Iranian bomb. But it took these experts numerous months-and hundreds of hours of labor-to equate each page, review it by hand for relevant material, and include that details into evaluations. With today's AI abilities, the very first two steps in that procedure might have been accomplished within days, possibly even hours, enabling analysts to comprehend and contextualize the intelligence rapidly.

Among the most intriguing applications is the way AI might change how intelligence is taken in by policymakers, allowing them to communicate straight with intelligence reports through ChatGPT-like platforms. Such capabilities would permit users to ask specific concerns and receive summarized, appropriate details from thousands of reports with source citations, assisting them make notified choices rapidly.

BRAVE NEW WORLD

Although AI provides numerous benefits, it likewise positions considerable brand-new risks, especially as adversaries develop similar technologies. China's advancements in AI, especially in computer system vision and surveillance, threaten U.S. intelligence operations. Because the country is ruled by an authoritarian regime, it lacks personal privacy constraints and civil liberty defenses. That deficit enables massive information collection practices that have yielded information sets of immense size. Government-sanctioned AI models are trained on huge quantities of individual and behavioral information that can then be utilized for different purposes, such as surveillance and social control. The presence of Chinese business, such as Huawei, in telecoms systems and software application around the globe could provide China with all set access to bulk data, significantly bulk images that can be utilized to train facial acknowledgment models, a particular concern in countries with large U.S. military bases. The U.S. nationwide security neighborhood should think about how Chinese models developed on such substantial information sets can give China a tactical benefit.

And it is not just China. The expansion of "open source" AI models, such as Meta's Llama and those created by the French business Mistral AI and the Chinese company DeepSeek, is putting powerful AI capabilities into the hands of users across the world at fairly budget friendly costs. A number of these users are benign, however some are not-including authoritarian regimes, cyber-hackers, and criminal gangs. These malign actors are using big language designs to rapidly generate and spread false and malicious material or to conduct cyberattacks. As seen with other intelligence-related technologies, such as signals intercept capabilities and unmanned drones, China, Iran, and Russia will have every incentive to share some of their AI advancements with customer states and subnational groups, such as Hezbollah, Hamas, and the Wagner paramilitary company, thereby increasing the risk to the United States and its allies.

The U.S. military and intelligence community's AI models will become attractive targets for adversaries. As they grow more powerful and main to U.S. nationwide security decision-making, intelligence AIs will become crucial national assets that need to be defended against enemies seeking to jeopardize or control them. The intelligence neighborhood should purchase developing protected AI and in establishing standards for "red teaming" and constant assessment to protect against prospective threats. These teams can utilize AI to imitate attacks, revealing prospective weaknesses and developing strategies to alleviate them. Proactive steps, consisting of cooperation with allies on and investment in counter-AI technologies, will be vital.

THE NEW NORMAL

These difficulties can not be wished away. Waiting too long for AI innovations to fully mature brings its own threats; U.S. intelligence capacities will fall back those of China, Russia, and other powers that are going full steam ahead in developing AI. To make sure that intelligence-whether time-sensitive cautions or longer-term tactical insight-continues to be an advantage for the United States and its allies, the country's intelligence community requires to adapt and innovate. The intelligence services need to quickly master using AI innovations and make AI a fundamental aspect in their work. This is the only sure way to ensure that future U.S. presidents get the very best possible intelligence assistance, remain ahead of their enemies, and protect the United States' delicate capabilities and operations. Implementing these changes will need a cultural shift within the intelligence neighborhood. Today, intelligence experts mainly construct products from raw intelligence and information, with some support from existing AI designs for voice and imagery analysis. Moving forward, intelligence officials should explore consisting of a hybrid technique, in line with existing laws, using AI models trained on unclassified commercially available information and fine-tuned with classified details. This amalgam of innovation and standard intelligence gathering could result in an AI entity offering instructions to images, signals, open source, and measurement systems on the basis of an incorporated view of regular and anomalous activity, automated images analysis, and automatic voice translation.

To accelerate the shift, intelligence leaders need to promote the advantages of AI combination, stressing the enhanced capabilities and efficiency it uses. The cadre of recently appointed chief AI officers has actually been established in U.S. intelligence and defense to function as leads within their companies for promoting AI innovation and removing barriers to the innovation's implementation. Pilot tasks and early wins can construct momentum and confidence in AI's capabilities, encouraging broader adoption. These officers can utilize the competence of national labs and other partners to test and improve AI designs, ensuring their efficiency and security. To institutionalise change, leaders should produce other organizational rewards, consisting of promotions and training chances, to reward innovative methods and those staff members and units that demonstrate efficient use of AI.

The White House has produced the policy needed for the usage of AI in nationwide security companies. President Joe Biden's 2023 executive order relating to safe, protected, and trustworthy AI detailed the assistance required to fairly and safely utilize the innovation, and National Security Memorandum 25, released in October 2024, is the nation's fundamental strategy for utilizing the power and handling the dangers of AI to advance nationwide security. Now, Congress will need to do its part. Appropriations are needed for departments and firms to produce the facilities needed for development and experimentation, conduct and scale pilot activities and assessments, and continue to buy examination capabilities to ensure that the United States is building reputable and high-performing AI technologies.

Intelligence and military neighborhoods are dedicated to keeping humans at the heart of AI-assisted decision-making and have actually developed the structures and tools to do so. Agencies will require guidelines for how their analysts should use AI designs to make certain that intelligence products fulfill the intelligence community's standards for reliability. The federal government will also require to maintain clear assistance for managing the data of U.S. citizens when it pertains to the training and usage of big language designs. It will be important to stabilize the use of emerging technologies with safeguarding the privacy and civil liberties of citizens. This implies augmenting oversight systems, upgrading pertinent frameworks to show the capabilities and dangers of AI, and promoting a culture of AI advancement within the national security apparatus that harnesses the potential of the innovation while safeguarding the rights and flexibilities that are fundamental to American society.

Unlike the 1950s, when U.S. intelligence raced to the forefront of overhead and satellite imagery by establishing much of the essential technologies itself, winning the AI race will require that neighborhood to reimagine how it partners with personal market. The personal sector, which is the main methods through which the federal government can realize AI progress at scale, is investing billions of dollars in AI-related research, information centers, and computing power. Given those business' developments, intelligence firms should focus on leveraging commercially available AI models and refining them with categorized data. This approach enables the intelligence community to quickly expand its capabilities without needing to go back to square one, enabling it to remain competitive with foes. A recent collaboration between NASA and IBM to develop the world's largest geospatial structure model-and the subsequent release of the model to the AI neighborhood as an open-source project-is an exemplary demonstration of how this kind of public-private partnership can operate in practice.

As the nationwide security neighborhood incorporates AI into its work, it must make sure the security and strength of its designs. Establishing requirements to release generative AI securely is crucial for maintaining the stability of AI-driven intelligence operations. This is a core focus of the National Security Agency's brand-new AI Security Center and its collaboration with the Department of Commerce's AI Safety Institute.

As the United States deals with growing competition to shape the future of the worldwide order, it is urgent that its intelligence agencies and military take advantage of the nation's innovation and leadership in AI, focusing particularly on large language models, to supply faster and more pertinent details to policymakers. Only then will they gain the speed, breadth, and depth of insight required to navigate a more complex, competitive, and content-rich world.