AI Race: Can Breakneck Speed Be Reconcile with Safety Imperatives?

A critique of AI safety from an OpenAI researcher directed at a competitor reveals a deeper industry struggle—a battle against itself.

It began when Boaz Barak, a Harvard professor currently on leave and working on safety at OpenAI, called the release of xAI's Grok model "completely irresponsible." His criticism wasn't about its headline-grabbing antics, but rather what was missing: a public system card, detailed safety evaluations, and the basic artifacts of transparency that have become fragile industry norms.

This was a clear and necessary call to action. However, a candid reflection from former OpenAI engineer Calvin French-Owen, published just three weeks after his departure, reveals the other side of the story.

French-Owen's account indicates that a significant number of OpenAI employees are indeed focused on safety, addressing very real threats such as hate speech, bioweapons, and self-harm. Yet he offers a crucial insight: "Most of the work being done isn't published," he wrote, adding that OpenAI "really should do more to get it out there."

Here, the simple narrative of a responsible actor reprimanding an irresponsible one falls apart. Instead, we see the industry's true dilemma laid bare. The entire AI sector is caught in what could be called the 'Safety-Velocity Paradox'—a deep, structural conflict between the competitive pressure to advance at breakneck speed and the ethical imperative to proceed with caution to keep people safe.

According to French-Owen, OpenAI operates in a state of controlled chaos. The company tripled its workforce to over 3,000 people in just one year, where "everything breaks when you scale that quickly." This chaotic energy is driven by the immense pressure of what he describes as a "three-horse race" toward artificial general intelligence, competing directly with Google and Anthropic. The result is a culture defined by incredible speed, but also by secrecy.

Consider the development of Codex, OpenAI's coding agent. French-Owen describes the project as a "mad-dash sprint," where a small team built a groundbreaking product from scratch in just seven weeks.

This serves as a textbook example of velocity—working until midnight most nights, and even through weekends, to make it happen. This is the human cost of such speed. In an environment moving this rapidly, is it any wonder that the slow, methodical work of publishing AI safety research feels like a distraction from the race?

This paradox does not stem from malice, but from a combination of powerful, interconnected forces.

There is the obvious competitive pressure to be first. There is also the cultural DNA of these labs, which originated as informal communities of "scientists and tinkerers" that prize groundbreaking innovations over systematic processes. Compounding this is a basic measurement problem: it's easy to track speed and performance, but exceptionally difficult to quantify a disaster that was successfully averted.

In today's boardrooms, the visible metrics of progress will almost always dominate over the unseen successes of safety. However, for the industry to move forward, the focus must shift from placing blame to fundamentally changing the rules of the game.

We must redefine what it means to launch a product, making the publication of a safety case as integral as the code itself. We need industry-wide standards that ensure no company is competitively disadvantaged for being diligent, transforming safety from an optional feature into a shared, non-negotiable foundation.

Above all, we must foster a culture within AI labs where every engineer—not just those in the safety department—feels a sense of responsibility.

The race to create AGI is not only about who gets there first, but about how we arrive. The ultimate winner will not be the company that is merely the fastest, but the one that shows the world that ambition and responsibility can—and must—advance together.

See also: Military AI contracts awarded to Anthropic, OpenAI, Google, and xAI

Want to learn more about AI and big data from industry leaders? Check out the AI & Big Data Expo held in Amsterdam, California, and London. This comprehensive event is co-located with other leading conferences, including the Intelligent Automation Conference, BlockX, Digital Transformation Week, and the Cyber Security & Cloud Expo.

Explore other upcoming enterprise technology events and webinars powered by TechForge here.