a16z's academy commits $840,000 per founding student before anyone pays tuition

About 50 students and $42 million comes to roughly $840,000 a head for the founding class ^[1]. The share earmarked for the students themselves is small: $50,000 in compute credits and a $5,000 travel budget each ^[4], which across 50 places is $2.75 million, or about 6.5 percent of the money ^[2]. Fortune does not say what the rest pays for.

The paying version arrives in 2028, and it starts with a regulator. The Academy intends to seek approval for a two-year program ^[5], and if it is approved Biyani expects to charge "something similar to elite private universities," which he put at $60,000 to $90,000 a year ^[6]. Hold the class near 50, run both years, take the $75,000 midpoint, and gross tuition is $7.5 million a year ^[3]. Biyani told Fortune that higher education is a $2 trillion industry ^[7]; $7.5 million is 0.0004 percent of it ^[5]. Recovering the $42 million at $75,000 a head takes 560 student-years, before costs ^[4].

The two accounts of the check do not match. Fortune reports $42 million from a16z and a group of individual investors, in a one-year program ^[1]. PYMNTS, citing the Wall Street Journal, reports the firm investing $35 million in a two-year, unaccredited school for students aged 16 to 22 ^[21]. The gap is $7 million ^[6]. If the $35 million is a16z's own money and the individuals supplied the balance, both figures hold, and neither publisher splits the total.

On Biyani's own description, the money buys time for students to pursue their own objectives. Teaching "is not the primary thing we plan on doing," he told Fortune. "The primary thing we're offering is the ability for students to pursue their own objectives." ^[11] Students will spend up to 80% of their time on projects, according to the Journal's account ^[22]. The ten founding partners are Anduril, Anthropic, Coinbase, Google, Meta, Nvidia, OpenAI, Palantir, Replit and Stripe, and their commitments "vary," Biyani said, adding that "Not all of them are going to take students for hiring." ^[13] Around 50 further hiring partners, among them Databricks and Boom Supersonic, are committed to interviewing or considering students and not to any number of hires ^[14].

In my view a16z has bought a year of proximity to about 50 selected teenagers and priced it as deal flow, and the 2028 tuition list is a second business that does not exist until a regulator signs. Biyani's counter is on the record: "We think there are lots of ways to turn this into a successful business," he told Fortune ^[8]. Bryan Caplan, the George Mason economist who wrote a book arguing college is mostly a credential, thinks it works at 50 and not above it. "Something that can work for 50 handpicked people isn't going to work even for the top 5,000," he said. "MIT is going to last for centuries. Don't worry about it." ^[19] The 5,000 he names is a hundred times the founding class ^[7].

It could go otherwise. A regulator could decline, leaving a program that is permanently free and permanently subsidised. Approval plus a full paying cohort would make the tuition business real. What I would look for is families writing cheques for $60,000 to $90,000 a year at a school with no accreditation, no degree and no federal student aid ^[2]. The third path is the co-ops, which the school's FAQ says are "typically" paid but not guaranteed ^[15].