The technology industry’s relationship with military applications remains complex, marked by internal dissent and public protests at giants like Google and Microsoft over contracts involving surveillance and defense.
Yet, Alexandr Wang, founder and CEO of influential AI company Scale AI, is charting a different course. Speaking at the Center for Strategic and International Studies (CSIS) in Washington on May 2, Wang offered a forceful defense of his company’s extensive work with the US Department of Defense, casting it not just as a business decision, but as a “moral imperative” essential for national security and geopolitical realities.
Wang argues that his perspective is shaped by unique experiences, including growing up in Los Alamos, New Mexico – the town synonymous with the Manhattan Project. “Knowing about the history of where I grew up, and the impact that Los Alamos and the Manhattan Project had on Pax Americana and the global order, it felt so clear that great AI technology needs to be applied to national security problems,” he explained at the CSIS event.
Observations during a trip to China, where he saw AI deployed for surveillance, further convinced him of the need for American leadership. “We’re at the brink of this incredibly powerful new technology, and the applications for national security are obvious,” Wang stated. “It’s going to be imperative for the US to stay ahead.”
From ‘Pariah’ to Pentagon Partner
This conviction led Scale AI, founded in 2016 to provide essential data labeling and curation for AI models, to pursue defense contracts early on. Wang acknowledged this was initially met with skepticism in Silicon Valley, stating previously, “We were a bit of a pariah.”
He added at CSIS, “One core belief I have is that over time, rationality prevails,” suggesting his view on the necessity of such work would become more accepted. Notably, Scale AI reportedly stepped in to take on data labeling work for the Pentagon’s controversial Project Maven initiative after Google withdrew in 2018 following employee protests.
Since then, Scale AI’s defense footprint has grown substantially through various government contract vehicles. The company secured a $249 million-ceiling Blanket Purchase Agreement (BPA) via the DoD’s Joint Artificial Intelligence Center (JAIC, now part of the Chief Digital and AI Office – CDAO) in January 2022.
An Other Transaction Agreement (OTA) awarded by the CDAO in August 2023 involves Scale AI providing its Data Engine for curating and annotating multimodal military data (video, radar, imagery). Furthermore, Scale AI was selected in February 2024 by the CDAO to help test and evaluate generative AI models for military use.
Its most prominent current project appears to be Thunderforge, a DIU program using AI agents (software designed to perform tasks autonomously or semi-autonomously based on goals) for military planning, focusing initially on US Indo-Pacific Command (INDOPACOM) and European Command (EUCOM), with partners including Anduril and Microsoft. An earlier, specific example includes a 2021 contract focused on accelerating USAF AI for intelligence, surveillance, and reconnaissance (ISR) data processing.
The Geopolitical Calculus
Wang consistently frames Scale AI’s defense work within the context of US competition with China. “The CCP has had an AI master plan,” he asserted at CSIS.
“They’ve discussed the need to win on the sort of implementation and proliferation of this technology throughout their military.” He has previously pointed to China’s hardware capabilities, claiming at Davos 2025 that DeepSeek possessed around 50,000 Nvidia A100 GPUs despite US export controls. These controls, which restricted Nvidia H20 sales to China effective April 15, 2025, were acknowledged by Wang as complex but perhaps insufficient, pointing to advancements from Chinese firms like DeepSeek and Huawei.
His stance aligns with views expressed in a January 2025 letter to President Donald Trump, where Wang urged greater US government investment in AI adoption – claiming China outspends the US 10-to-1 – and cautioned against regulations that might hinder American innovation. “Attracting the best technology companies to flourish in America relies on the right regulatory framework,” he wrote, according to The National.
An Industry Navigating Controversy
While Wang’s vocal advocacy is distinct, Scale AI operates within an industry increasingly intertwined with defense, albeit one still navigating ethical lines and internal pressures. Palantir and Anduril Industries, both heavily involved with the DoD, formed a consortium in December 2024 to improve military AI data processing.
Anthropic partnered with Palantir and AWS in November 2024 to deploy its Claude models for intelligence use. OpenAI, a Scale AI client, updated its policies in January 2024 regarding military use, later adding former NSA director Paul Nakasone to its board and partnering with Anduril on counter-drone tech.
Even Google, despite reversing its earlier ban on military AI work in February 2025, faced renewed internal pressure from DeepMind staff in August 2024 over Project Nimbus and later fired dozens involved in protests. Microsoft experienced similar disruptions and firings in April after employees protested Azure AI contracts with the Israeli military, interrupting AI CEO Mustafa Suleyman.
One protester shouted, “Stop using AI for genocide, Mustafa… You have blood on your hands.” Suleyman responded, “thank you for your protest. i hear you.” Wang himself has faced public disagreement, such as audience pushback at Web Summit Qatar in February 2025 regarding his competitive framing. Despite these industry crosscurrents, Wang remains steadfast in his view that contributing to national security through AI is essential.
