In the closing minutes of a recent Motley Fool Money episode titled “The Old Software Moat Is Dead,” host Adam Field, chief AI officer at Tungsten Automation, offered the sharpest framing yet of the two forces pulling enterprise AI in opposite directions. His answer to the question of what excites and worries him most about the technology came down to a single tension: the same models that can teach a child in a village with no school can also be turned into weapons against the software supply chain that runs the global economy.
Field is worth listening to on this because of his vantage point. Tungsten Automation serves more than 25,000 organizations, including roughly 40% of the Fortune 100, which puts him inside the workflows where enterprise AI either delivers ROI or quietly burns budget. His conversation with analyst Rachel Warren focused on how the legacy software moat has effectively collapsed, and the closing segment turned to what happens next.
The Risk: Cybersecurity and the Open Source Problem
Field pointed to cybersecurity as his sharpest concern, citing bad actors injecting nefarious code into open source tools and referencing the controversy around the Mythos and Fable collaboration with Anthropic and the US government, which was reportedly halted after being deemed a potential cyber weapon.
That reference is worth unpacking. According to Anthropic’s own statement, the US government issued an export control directive on June 12, 2026, suspending access to Fable 5 and Mythos 5 for all foreign nationals, citing national security authorities. Anthropic complied by disabling the models for all customers, while disputing that the underlying jailbreak finding justified pulling a commercial model deployed to hundreds of millions of users. The episode was a rare case of Washington treating a frontier model like a controlled munition, and it validates Field’s broader point that cybersecurity risk is now a first-order variable in enterprise AI deployment decisions.
The federal budget is moving in the same direction. The FY 2027 Department of War budget requests $58.5 billion for artificial intelligence, including $46.0 billion for a sovereign AI Arsenal, with an explicit focus on defending critical infrastructure and the defense industrial base against malicious cyber attacks. Enterprise buyers evaluating AI vendors are increasingly asking the same questions the government is: who controls the model weights, where does training data flow, and how quickly can access be revoked?
The Promise: Multiplying a Teacher
On the upside, Field described hearing an education expert with decades of experience talk about how tools like ChatGPT and Claude are giving people in impoverished parts of the world access to quality education that was previously out of reach, comparing it to multiplying a teacher “exponentially” to reach underserved populations. He paired that with a concrete personal productivity point: tasks that once required tracking down the right book or website now take a fraction of the time.
That framing matters for investors trying to separate durable AI value from hype. The clearest ROI stories so far involve automating knowledge work at scale, whether it is Helport AI (NASDAQ: HPAI) validating an AI labor workflow across the consumer finance lifecycle or authID (NASDAQ: AUID) deploying biometric authentication for a global retailer with more than 100,000 employees. Both are enterprise deployments where the unit economics are legible, which is the signal Field spent most of the episode urging investors to look for.
What to Watch
Field’s dual framing lands at a moment when capital is still flooding in. Server backlogs have hit $57 billion, and Bill Ackman has publicly bet on a $700 billion hyperscaler AI spending wave. The winners will likely be the vendors whose products survive both tests Field described: measurable enterprise productivity gains on one side, and defensible security posture on the other. Investors should keep an eye on how quickly companies can produce hard numbers on both.
Contact [email protected] for any questions or corrections.