Jasper AI: A Dilemma of a Thin Wrapper

GPT's Blessing and Curse on The Unicorn Turned Underdog

To use, or not to use OpenAI’s API (e.g. GPT-3) as a foundation for your business? Jasper, an AI copywriting startup, might just be the one learning this lesson the hard way right now. It soared to fame in 2021, only to be quickly outshone by ChatGPT. It’s still considered a member of the Generative AI (GenAI) Unicorn club, but its standing is unstable: a few weeks ago, it saw a $0.3 billion drop in valuation. It makes it a prime case to study, being one of the largest enterprises initially wholly reliant on GPT-3 model. Today, we'll examine the insights this case offers.

  1. How to survive ChatGPT?

  2. The starting point of Jasper

  3. No doomism in Jasper’s mission

  4. Financial situation

  5. How does Jasper (plan to) make money? 

  6. What does it offer?

  7. Tech behind Jasper

  8. Bonus: All important links about the founders

How to survive ChatGPT?

November of 2022, OpenAI released ChatGPT giving the public power of Jasper and similar tools in a user-friendly interface making it everyone’s favorite in an instant. While the AI tool managed to hypnotize the public with its magic, it caused mayhem among the industry insiders; including Jasper’s CEO who had some urgent calls to make.

In a virtual meeting, Dave Rogenmoser, Jasper’s CEO, found himself on one side of the screen, while Sam Altman, OpenAI’s CEO, appeared on the other. The brainchild of Rogenmoser, had been built upon OpenAI's flagship model, GPT-3, establishing a close-knit relationship between the two tech startups. The Information reported that they also share a Slack channel for exchanging updates and feedback related to GPT-3.

Rogenmoser's attention was piqued by the release of ChatGPT, a product strikingly similar to Jasper's, but… free of charge. 

Considering the significant payments made by Jasper, a Y Combinator-backed enterprise, for the use of GPT-3, Rogenmoser felt a need to address the matter with OpenAI. While recognizing that OpenAI had no obligation to Jasper, the CEO expressed the necessity for a reconsideration of their business strategies if Silicon Valley’s new darling intended to continue providing its AI chatter box for free.

Soon, ChatGPT indeed introduced a monthly fee… significantly lower than Jasper’s prices.

That became a huge problem for many so-called 'thin wrappers' that built their products around OpenAI technology.

In September 2023, less than year after making that call to Altman, Rogenmoser has decided to step down as the CEO to help the company gain more enterprise customers through its next chief, Timothy Young, previous president of Dropbox Inc.

In October 2023, the company released an end-to-end AI copilot to personalize each company and combine business context and memory with purpose-built skills. With this launch, the company also added analytics and AI insights for better performing content, an intelligence hub to securely tap into brand, strategy and audience information, and campaign acceleration tools to monitor processes.

Jasper’s path was never rosy; at some point it may have seemed like an instant triumph, yet its origins trace back to a lengthy eight-year journey of trial and error for its co-founders.

The starting point of Jasper

In 2014, Rogenmoser, JP Morgan (yes, his name indeed matches the financial giant's), and Chris Hull forged a pact, vowing to establish a company, generate $30,000 per month, and uphold their camaraderie above all else.

Following the closure of two unsuccessful Software as a Service (SaaS) ventures, the trio eventually landed up with a digital marketing agency, which trajectory soon shifted towards educational courses, and understanding the intricacies of marketing. But there was no scale, and much more hassle than money. Dissatisfied with their agency and info-publishing ventures, Dave sought a software business with recurring revenue. Over a weekend, he and co-founders created a widget displaying recent course purchases, doubling their sales. Testing it on friends' websites yielded positive sales conversions, leading to the inception of Proof. This led them to the Y Combinator program.

At some point Dave went back to teaching companies how to do marketing and was struggling with their inability to write a good marketing copy. Around that time in 2020, he started to see conversations on Twitter about GPT-3. He thought that AI might help with his struggles.

“In late 2020, OpenAI set up a small, private beta for developers to see how they might use the technology in the real world,” reports The Information. The beta was extremely limited, but Rogenmoser used his Y Combinator connections to get access that December.” That gave him a huge head start in developing a product on this technology.

Morgan, who self-taught himself to code, created a front-end and in January 2021, Rogenmoser, Morgan, and Hull introduced their brainchild, originally dubbed Conversation AI, later undergoing a name change from Jarvis, inspired by the AI butler in "Iron Man," to Jasper, following a legal altercation with Marvel over the use of the former name.

For Jasper, creating a product that resonates with customers was the objective. Rogenmoser recalled the moment he realized they were on the right course, citing instances of customers tattooing Jasper's signature logo – a cheerful robot – on their limbs. 

No doomism in Jasper’s mission

Jasper was one of the first to start utilizing the power of generative AI, immediately recognizing it as a tool. The founders’ aim became to democratize content generation, enabling users from diverse professional backgrounds to produce content efficiently. This objective speaks to the core of Jasper's ethos – to serve as a tool that complements human intellect rather than replace it.

Also, it aligns with Rogenmoser's objective of earning money on a recurring basis. He understood: time is money, which saved the company from long-term AI risk considerations.

What the company does pay attention to is echoing concerns regarding biases stemming from data. Explaining the importance of customer feedback in mitigating biased content, the website highlights the necessity for generative AI startups to remain attuned to such issues and adjust their training data accordingly.

Financial situation

We’ve put together a concise explanation for your convenience, available only for Premium subscribers →subscribe here

Dave Rogenmoser, Co-founder, ex-CEO

 JP Morgan, Co-founder, Board Member, ex-CTO

Chris Hull, Co-founder

Timothy Young, CEO

Conclusion

Jasper's journey sheds light on the intricate dynamics and inherent hurdles of building a business on external AI frameworks like OpenAI’s GPT-3. It's a cautionary tale showing how quick technological advancements can disrupt even well-funded and seemingly stable AI enterprises. Despite being a pioneer in democratizing content creation, overnight Jasper found itself wrestling with competitive pricing and product differentiation post ChatGPT's entry. The move to a proprietary AI engine in 2023 reflects a strategic shift towards self-reliance and specialization amidst a crowded and volatile market. The unfolding Jasper-ChatGPT narrative begs the question: Can 'thin wrappers' around existing AI technologies endure in a rapidly evolving landscape, or will they be overshadowed by the very frameworks they rely upon?

You are currently on the free list. Join Premium members from top companies like Datadog, FICO, UbiOps, etc., AI labs such as MIT, Berkeley, and .gov, as well as many VCs, to learn and start understanding AI →

Thank you for reading, please feel free to share with your friends and colleagues. In the next couple of weeks, we are announcing our referral program 🤍

Join the conversation

or to participate.