‘OpenAI Files’ Push for Accountability in the Race to AGI

As OpenAI and its CEO Sam Altman continue to position artificial general intelligence (AGI) as a near-future reality, a growing chorus of critics is pushing back—not on the technology itself, but on how it’s being built, governed, and commercialized.

That concern is the foundation of The OpenAI Files, a public archive launched by two nonprofit watchdogs: the Midas Project and the Tech Oversight Project. Framed as a call for transparency and accountability, the Files aim to document what the groups describe as “concerns with governance practices, leadership integrity, and organizational culture” at OpenAI.

Their broader goal, according to the site’s “Vision for Change,” is to ensure that leadership in AGI development reflects the weight of the challenge—and the potential consequences. “The companies leading the race to AGI must be held to, and must hold themselves to, exceptionally high standards,” the statement reads.

From Promise to Profit

The Files offer a detailed account of OpenAI’s shift from its original nonprofit mission to its current investor-driven structure. When it was founded, OpenAI had committed to a capped-profit model—limiting investor returns to 100x, with the rest of the value flowing to humanity. That cap has since been removed, a change OpenAI said was necessary to secure continued investment.

The pressure to scale quickly has led to other compromises, according to the archive. The Files criticize OpenAI’s safety testing as rushed and point to a “culture of recklessness,” fueled by the need to ship products rapidly and meet growing investor expectations. That same pressure, watchdogs argue, has led to controversial practices like scraping data without consent and building infrastructure that draws heavily on local energy grids—sometimes contributing to power shortages and higher consumer costs.

Leadership, Conflict, and Control

Beyond structural changes, the Files also question the integrity and decision-making of OpenAI’s leadership. The archive highlights potential conflicts of interest among board members and includes a list of startups reportedly tied to Altman’s personal investment portfolio—some of which operate in areas that overlap with OpenAI’s own product lines.

Tensions around leadership came to a head in 2023, when OpenAI’s senior staff attempted to remove Altman. One of those voices, former chief scientist Ilya Sutskever, reportedly said, “I don’t think Sam is the guy who should have the finger on the button for AGI.”

The Files argue that this kind of concentration of power—where decisions about world-changing technologies rest with a small, largely unaccountable leadership group—underscores the need for outside scrutiny.

Why the Focus on OpenAI?

While the OpenAI Files raise concerns that could apply broadly across the AI industry, their focus is squarely—and intentionally—on OpenAI. The reasons for that are rooted in how the company positioned itself, how it has evolved, and how much influence it holds in the race to AGI.

1. OpenAI Positioned Itself as the Moral Actor: OpenAI was founded as a nonprofit with the stated mission of building AGI “for the benefit of humanity.” It publicly committed to safety, transparency, and capped profits. That ethical framing set it apart from competitors—but also raised expectations.

Critics argue that violating those early commitments is more consequential than never having made them at all. The perception is that OpenAI claimed the high ground, then compromised it to scale rapidly and secure funding.

2. OpenAI Is Leading the Race—and Setting Precedents: Sam Altman and OpenAI are widely seen as pacing the AGI race. Their high-profile model launches, strategic partnerships (especially with Microsoft), and public visibility give them outsized influence. The Files suggest that when a company with that kind of reach makes opaque or controversial decisions, others may follow.
   

3. Leadership and Governance Questions Are Specific: Unlike many competitors, OpenAI has experienced visible internal instability—most notably the attempted ousting of CEO Sam Altman in 2023. The Files highlight reported concerns from insiders like Ilya Sutskever, along with questions about board-level decision-making and potential conflicts of interest. That level of internal scrutiny simply hasn’t emerged as clearly from companies like Google DeepMind or Anthropic.

Vision for Change

Before outlining their recommendations, the Files lay out four core values they believe should guide AGI development:

• Integrity in Leadership – Leaders of AGI projects should demonstrate ethical behavior, humility, and transparency.

• Accountability in Governance – AGI development must include external oversight, robust safety practices, and clear consequences for misconduct.

• Commitment to Shared Benefit – The benefits of AGI should be broadly distributed, not concentrated among a few individuals or investors.

• Transparency and Trust – Companies must earn public trust through openness about their structures, decisions, and goals.

These values are intended to serve as a blueprint—not just for OpenAI, but for any organization claiming to shape humanity’s future with AGI.

What the Files Propose

The OpenAI Files are not just a critique of leadership—they’re a call for systemic reform. The organizations behind the project argue that companies developing AGI must be held to higher standards of governance, transparency, and accountability.

Specifically, they recommend:

• Stronger oversight structures that match the scale and impact of AGI-level development

• Clear disclosures of investor influence and leadership conflicts of interest

• Renewed commitment to public benefit, especially for companies that began with nonprofit or mission-driven foundations

• Greater public awareness of how AGI is being shaped, and by whom

But the Files also raise a broader tension: while their concerns are serious and well-documented, the focus is almost entirely on OpenAI. That narrow scope has drawn criticism of its own. If the goal is to ensure AGI is developed in the public interest, shouldn’t the same scrutiny apply to every company in the race—including Google DeepMind, Anthropic, Meta, and others?

By targeting only one organization, the watchdogs risk reinforcing the perception that this is more about one company’s leadership than about industry-wide accountability. And yet, the underlying message remains urgent: as AGI development accelerates, the public deserves more visibility into the structures guiding that future.

Can Oversight Happen Without Willing Regulators?

One of the central tensions exposed by the OpenAI Files is this: even if greater oversight is urgently needed, who’s in a position to provide it?

So far, both the U.S. government and many leading AI companies have resisted binding regulation. The current administration has promoted voluntary safety commitments and public-private partnerships, but has stopped short of proposing enforceable laws that would limit corporate power or require structural transparency.

At the same time, companies like OpenAI, Google, and Anthropic have publicly supported general safety principles—but have been far less enthusiastic about external scrutiny of their leadership or investor influence.

This raises a critical question: if neither government nor industry is demanding stronger oversight, where will it come from? Watchdog projects like the OpenAI Files aim to fill that gap—but without enforcement power, their influence depends on public pressure, media attention, and internal dissent from employees and researchers.

Calls for accountability, no matter how urgent, can only go so far without structures to enforce them. As AGI development accelerates, the debate is no longer just about technical safeguards—it’s about power, governance, and who gets to decide the terms of the future.

What This Means

The OpenAI Files are less about stopping AGI development and more about demanding that it unfold with transparency, ethics, and long-term public interest in mind. By documenting structural changes, leadership dynamics, and investor influence, the Files attempt to shift the focus from inevitability to accountability.

In a field defined by speed and scale, this project raises the question of whether public trust can keep pace with private ambition. As more AI companies race toward AGI, the debate may no longer be about if they’ll get there—but how, and at what cost.

Editor’s Note: This article was created by Alicia Shapiro, CMO of AiNews.com, with writing, image, and idea-generation support from ChatGPT, an AI assistant. However, the final perspective and editorial choices are solely Alicia Shapiro’s. Special thanks to ChatGPT for assistance with research and editorial support in crafting this article.