More AI Nonprofits Are Launching, But Don’t Expect OpenAI-Type Drama

Given all the controversy OpenAI has generated as a capital-intensive startup founded as a nonprofit, one might think AI-focused founders today would opt to avoid that governance structure.

In reality, however, the number of nonprofit AI-focused startups is actually growing at a fast clip. Nonprofit accelerator Fast Forward said that this year more than a third of applicants for its latest class were AI companies. They’re applying the technology in areas including education, healthcare and critical infrastructure.

As we observed a few months ago, tech nonprofits aren’t actually that unusual. While most entrepreneurs do choose for-profit business models, some don’t find it the right fit. This includes prominent names such as Wikipedia, Mozilla, and Khan Academy, which operate as nonprofits.

As OpenAI’s drama continues to unfold, we thought it would be timely to touch base with Fast Forward co-founder Kevin Barenblat about lessons learned and how the next crop of AI-focused nonprofits is likely to scale.

OpenAI looks like a unique case

Per Barenblat,OpenAI is a unique case. It’s unprecedented for a for-profit startup to get a $10 billion financing offer from Microsoft — let alone a nonprofit (albeit one with a for-profit subsidiary.) It’s also unusual to see an ostensibly nonprofit company develop technology that’s seen as having such high commercial value.

“The idea when they started was that AI was both an opportunity and a threat,” he said. And the idea of the nonprofit was that the benefits would accrue to humanity and would not be about maximizing profit.

What happened a number of years ago, he observed, is that the mission became very expensive. “They decided the path would require an amount of capital that would be hard to raise inside a nonprofit. And that’s how we ended up where we are.”

Newer AI nonprofits

Today, there are scores of AI-focused nonprofits, most of which expect they will neither need nor be capable of raising billions to scale.

Rather than focus on developing core enabling technologies, many are working off existing generative AI platforms. For example Playlab, a Fast Forward accelerator participant focused on AI-enabled tools for teachers and students, lets users choose from Anthropic or OpenAI models to build learning experiences.

On the healthcare side, meanwhile, Boston-based RebootRX has raised at least $1 million for its mission of fast-tracking affordable cancer treatments using repurposed generic drugs, AI technology, and innovative funding models. 

Another nonprofit, India-based Karya, extends jobs to rural citizens who provide language data for automatic speech recognition systems.

More broadly, across the globe dozens of organizations, including nonprofits, academic groups, and research centers, are dedicating work around ethical approaches to AI. Some that came up in our perusal include:

• AlgorithmWatch, based in Berlin and Zurich, is focused on advocating for algorithms and artificial intelligence that strengthen “justice, democracy, and sustainability.”
• JoyEducation, a virtual clinic for students with reading difficulties, incorporates AI voice technology to tailor its curriculum.
• Earth05, an upstart nonprofit out of Barcelona, is looking to deploy AI toward global water conservation efforts.

Business models in an age of AGI

While we may (or may not) be a ways away from achieving artificial general intelligence, it’s apparently not too soon to be worried about what kinds of business endeavors it will power.

As with virtually any transformative technology, getting into the wrong hands — or even those of well-meaning but misguided enablers — could have disastrous consequences. As such, it’s not surprising to see a groundswell of founders attempting to direct AI’s promise to helpful purposes.

Illustration: Dom Guzman