Databricks, Perplexity co-founder pledges $100M on new fund for AI researchers

Andy Konwinski, computer scientist and co-founder of Databricks and Perpelexity, announced on Monday that his personal company, SheForms a new research institute on supported AI with a commitment of $ 100 million in its own money.

Laude Institute is less a research laboratory on AI and more as a fund that seeks to make structured investments similar to subsidies. In addition to Konwinski, the Institute’s board of directors includes the professor of UC Berkeley, Dave Patterson (known for a series of award-winning research), Jeff Dean (known as Google chief scientist) and Joelle Pineau (Vice-President of IA research).

Konwinski announced the institute’s first subsidy and “flagship” of $ 3 million a year for five years, and he will anchor the new AI systems laboratory in UC Berkeley. This is a new laboratory led by one of the famous Berkeley, Ion Stoica, current director of Sky Computing Lab. Stoica is also co-founder of Anyscale Startup (an AI and Python platform) and the Big Data Company data company, both developed in the Berkeley laboratory system.

The new AI systems laboratory should open its doors in 2027 and, in addition to Stoica, will include a number of other well -known researchers.

In his blog Announcing the institute, Konwinski described its mission as “built by and for computer researchers … We exist to catalyze the work which does not simply push the ground forward but the guide towards more beneficial results.”

It is not necessarily a direct excavation in Openai, which started as a research installation on AI and is now, no doubt, consumed by its enormous commercial side. But other researchers also fell prey to the appeal of money.

For example, the popular era of AI researchers Faced with controversy When he revealed that Optai supported the creation of one of his AI references which was then used to unveil its new O3 model. The founder of Epoch also launched a startup with a controversial mission Replace all human workers everywhere with AI agents.

Like other AI research organizations with commercial ambitions, Konwinski has structured his institute through borders: as a non -profit organization with an operational arm of the public benefits company.

He divides his investments in two buckets which he calls “slides and moon strokes”. Flows are intended for starting phase research which can benefit from subsidies and practical assistance. The blows of the moon are, as its name suggests, for “laboratories on a long horizon take up challenges in terms of species such as AI for scientific discovery, civic discourse, health care and reskilling of the workforce”.

His laboratory has, for example, collaborated with “Terminal-Bench”, a reference led by Stanford for the way in which AI agents manage tasks, Used by Anthropic.

One thing to note, the Konwinski company, Laude, is not only a research-writing institute. He also co-founded a for-profit venture capital fund launched in 2024. The co-founder of the fund is the former VC Pete Sonini of the NEA VC. As Techcrunch previously reported it, Laude conducted an investment of $ 12 million in the IA Arcade infrastructure startup. He also quietly supported other startups.

A Laude spokesperson tells us that even if Konwinski has promised $ 100 million, he is also looking for and opens up investments from other successful technologists. As for the way Konwinski has raised sufficient fortune to guarantee $ 100 million for this new company: Databricks closed $ 15.3 billion The financing tour in January which appreciated the company at $ 62 billion. Perplexity last month secure Also an assessment of $ 14 billion.

Does the world really need another research on the AI ​​”good for humanity” or with a disnteited non-profit / commercial structure? No, and yes.

AI research has become more and more muddy. For example, AI benchmarks designed to prove that the model of a particular supplier works better has become abundant these days. (Even dirty To his own reference LLM for CRMS.)

An alliance which includes tastes of Konwinski, Dean and Stoica supporting truly independent research which could one day turn into independent and thickening human trade could be an attractive alternative.