OpenAI, in one of its previous announcements in 2019, has revealed its transition from a nonprofit entity to a "capped-profit" organization, which places a limit on returns from investments beyond a certain limit. This change, and the manner in which it was carried out, have left some apprehensive that OpenAI may start to resemble any other AI start-ups in the industry.
Hereafter, any profits derived from investments in the OpenAI LP (Limited Partnership), will be channeled into a supervising nonprofit company. This body will then distribute and raise capital at its discretion. The cap comes into effect when the profits surpass a 100x return.
In layman's terms, if one were to invest $10 million today, the profit cap of OpenAI LP would only be triggered once that $10 million accumulates $1 billion in returns. This is why there's a growing concern that this "limited" structure is merely nominal. Furthermore, there can be repercussions on its equity compensation model.
What is the Reason Behind This?
OpenAI LP outlined the reasoning behind its decision in a blog post.
“We’ll need to invest billions of dollars in upcoming years into large-scale cloud computing, attracting and retaining talented people, and building AI supercomputers.
We want to increase our ability to raise capital while still serving our mission, and no pre-existing legal structure we know of strikes the right balance. Our solution is to create OpenAI LP as a hybrid of a for-profit and nonprofit—which we are calling a “profit capped” company.”
In essence, the company is acknowledging that, as a nonprofit, it struggled to garner the necessary funding to meet its objectives— a scenario that investors, who typically anticipate immediate returns, can find discouraging. Although profits can be made from offshoots and subsidiary businesses, injecting capital into a nonprofit doesn't usually promise significant financial gains.
The financial constraint would not pose as much of an issue if OpenAI LP did not find itself in a competitive landscape alongside industry giants like Google and Amazon, all vying for experts in artificial intelligence and cloud computing. Furthermore, the steep cost of development exacerbates the situation.
Similar challenges existed, albeit to a lesser extent, back in 2015 when OpenAI LP was established. However, as the artificial intelligence founders then noted in relation to AI systems:
“Our goal is to advance digital artificial intelligence in the way that is most likely to benefit humanity as a whole, unconstrained by a need to generate a financial return. Since our AI research is free from financial obligations, we can better focus on a positive human impact.”
It’s safe to say that there can’t be any interpretations for this.
That being said, OpenAI LP isn't a pioneering nonprofit to grapple with financial constraints; it's an indisputable reality that competing financially with multinational giants is a Herculean task, especially in an industry where budget significantly influences success. One could interpret their decision as viewing profitability as a means to alleviate financial burdens. This is a perspective worth considering along with human values intertwined with machine learning and AI systems.
The novel arrangement sees OpenAI LP carrying out the artificial intelligence core activities the firm is recognized for, which include engaging in intriguing and potentially broadly relevant AI systems research, sometimes reserved to safeguard global interests. And let’s not deny it, ChatGPT is actually taking the tech industry and other relevant sectors by storm when it was first introduced. It hasn’t lost its steam since.
However, the operations of the LP will be overseen by OpenAI Inc., also referred to as OpenAI Nonprofit (this structure will be detailed further). Any profits generated by the LP that exceed the 100x multiplier will be channeled to the nonprofit. This capital will then be deployed to administer educational initiatives and conduct advocacy efforts.
The private company defends its seemingly substantial profit "cap" by suggesting that if it can successfully develop a functioning artificial general intelligence (AGI, a concept that, while somewhat ambiguously defined, is arguably the ultimate objective of current AI research), “we expect to generate orders of magnitude more value than we’d owe to people who invest in or work at OpenAI LP.”
Nonetheless, the 100x figure seems to represent a significant increase. It's conceivable that similar objectives could have been attained with a 10x or 20x multiplier, allowing for substantial returns without making near-term profits seem practically boundless. Future investment rounds will actually feature a reduced multiplier; the current one is designed to incentivize investors who are open to assuming a slightly higher degree of risk.
How Does It Impact Employee Equity Compensation?
The OpenAI transitioned proposal is twofold: a fixed base salary and what they refer to as 'Profit Participation Units' or 'PPUs'. The base salary is a straightforward, regular monetary compensation received biweekly.
The complexity kicks in with the profit participation units. Their application can be perplexing, especially considering that different firms may use profit participation or profit interest grants in varying ways.
Profit Interest Units (PIUs) form the basis of OpenAI's Profit Participation Units (PPUs) generative models and its AI's general intelligence when used with platforms, such as the Bing search engine. PIUs are a customizable form of equity compensation common among private equity-backed startups.
In simple terms, owning a PIU entitles you to a share of the profits proportionate to the total number of PIUs. For instance, owning 1 PIU out of 100 means you're entitled to 1% of the company's profits.
If the private company isn't profitable, a held PIU won't yield any immediate value. However, it could still hold potential value to someone expecting future profits.
Unique to profit interests, they often come with a 'liquidation threshold' equal to the company's equity value. The company must reach profits at or above this threshold for the PIU to hold value. Recipients receive the units upon vesting at no extra cost, offering a tax advantage as they are tax-free upon issuance and vesting, with capital gains tax applicable only upon profit receipt or sale.
In contrast, with a Restricted Stock Unit (RSU), typical in larger companies, employees are taxed immediately upon vesting since RSUs, a percentage of company equity, hold a value based on the company's current stock price.
How PPUs at OpenAI Work?
In their job offers, OpenAI nonprofit provides an estimated value of the Profit Participation Units (PPUs) being granted. These PPUs vest over four years, at a rate of 25% per year. Unlike stock options, employees are not required to buy PPUs. They all have an equal value, and during a tender offer, accredited investors buy PPUs directly from employees. OpenAI nonprofit values PPUs according to the latest price investors paid for employee-held PPUs.
However, at the offer stage, candidates are not informed about the total number of PPUs they will receive or the total number in existence, making it unclear what percentage of profits they're entitled to. While this lack of transparency is common among non profit startups, it's generally viewed as poor practice and disadvantageous to employees. Moreover, there are no guarantees for tender offers, and their frequency is not predetermined.
PPUs come with a 2-year lock, prohibiting new hires from selling their units within the first two years. Also, growth is capped at 10x the original value. Therefore, if an offer includes $2M worth of PPUs, the maximum they could be sold for would be $20M.
OpenAI nonprofit adheres to an internal compensation guideline, making negotiation difficult according to some non profit sources. Existing public data supports this, showing similar base salaries and PPUs among employees. OpenAI uses an internal tool to appraise the value and growth of PPUs, hence, specific numbers aren't likely public knowledge.
The number of PPUs offered traditionally is unknown, preventing specific unit valuation. However, venture capitalists have roughly valued the company between $27 billion and $29 billion.
It's crucial to remember that PPUs hold no redeemable value if OpenAI doesn't generate profits. Their non profit value derives from the willingness of investors to purchase them.
Some candidates have reported OpenAI's expectations of turning a profit only upon achieving Artificial General Intelligence (AGI). But just like with any other startup, there are potential risks with equity, including the possible worthlessness of the units.
Equity Plan Advice for Tech Startups
For startups in tech, like OpenAI, adopting a Restricted Stock Units (RSUs) approach can be an advantageous move. One of the main appeals of RSUs is their simplicity. Employees receive a certain number of units, which translate into shares at a predetermined vesting schedule. This uncomplicated structure makes it easier for employees to comprehend their equity package and its potential value, eliminating the complexity associated with terms like 'profit participation units' or 'liquidation thresholds'.
Transparency is another notable advantage of RSUs. At the time of the grant, employees are aware of the number of units they're receiving, which become actual shares of the company when vested. This transparency makes it straightforward for employees to calculate their stake in the company and understand the impact on their earnings if the company performs well. Hence, for tech startups aiming to attract talent while ensuring their employees have a clear understanding of their compensation, RSUs can be an excellent choice.
To find out more about RSUs and how they can be managed efficiently, feel free to check out our blog.
.png){kind=logo}