The grand goal of OpenAI, according to head of sales Aliisa Rosenthal, is to “ensure that artificial general intelligence (AGI) benefits all of humanity.” This goal is fully aligned with her team’s daily efforts in sales, marketing, partnerships, and customer success, which include deploying ChatGPT Enterprise at more than 260 businesses and 150,000 employee users.
Although there isn’t a single definition for artificial general intelligence (AGI) in the AI industry, Rosenthal, who reports to OpenAI COO Brad Lightcap, recently called her 150-person go-to-market team “AGI sherpas” who “help our customers and our users transition to the paradigm shift of AGI” in an interview with VentureBeat. She also made no secret of the fact that ChatGPT’s initial purpose was only to collect additional data for GPT-3.5 and that OpenAI leverages user comments to enhance its models toward AGI.
The OpenAI sales and go-to-market team is expanding rapidly; more than twenty opportunities are listed on the company’s Careers page. Rosenthal stated that the team does not operate under quotas or commissions. The power of GPT and DALL-E was “magical,” according to Rosenthal, who says she took a chance by joining OpenAI two years ago following a four-year tenure as VP of sales at SaaS company WalkMe — months before ChatGPT led to the current generative AI boom. She said she knew there had to be a business application and was “pretty determined to find out what that was.”
Rosenthal also talked about her initial conversation with OpenAI CEO Sam Altman, her father’s foresighted forecast regarding OpenAI, and her team’s sherpa emoji in the interview, which has been trimmed and shortened for clarity.
Sharon Goldman: You recently shared on LinkedIn that the OpenAI sales team was once a “tiny rag-tag team,” but that number has since increased to over 100 salespeople. For you, how bizarre was that evolutionary arc?
Aliisa Rosenthal: There were only about 15 people working in this go-to-market company when I started two years ago. When ChatGPT first started, there were probably about thirty of us. As you may imagine, we had a lot of interest in ChatGPT and have been expanding the staff as quickly as possible over the past year. There are currently about 150 of us.
Where Have The Times During This Trip When You Thought, “Wow, I Need To Take A Minute To Calm Down” Occurred?
AI was mostly limited to experimental teams, beta projects, and little firms when I first started working here. Despite being a consumer product, ChatGPT has captured the interest of boardrooms worldwide, which is what’s made it so fascinating. All executives seemed to be asking themselves, “How do we start implementing AI in our business processes and the products that we build? ” at that point.We felt, “Okay, it’s time to really invest in and build our ability to support these enterprises as they’re asking these questions,” at what was clearly a key juncture in an otherwise unanticipated transition.
The introduction of our ChatGPT enterprise package marked the second significant milestone. With this, we actually have a product market fit, so it’s been fantastic to launch and start creating a customer success team around it. After our early September launch, we swiftly added hundreds of businesses as enterprise clients. That was yet another turning point in the climbing journey.
How Did Your Initial Meeting With Sam Altman, The CEO, Go?
I’ve been working in IT for about 20 years. I’ve held positions at several quick-growing startups. Additionally, I worked at an AI firm for almost ten years, which helped me have a little more understanding of neural networks and cluster analysis than I could have otherwise had when I tested out GPT-3 for the first time in December 2021.
I had contacted OpenAI because I knew someone who worked there. He put me in touch with Sam, who was rather open with me, admitting that he had no idea what to expect from sales here. It most likely won’t resemble anything else. It most likely will differ slightly.
The boldness of his mission and the organization’s objectives really struck me. I made a major mistake. I genuinely make jokes about how all of my buddies in venture capital warned me not to accept this role when I accepted the offer. They advised going to a place where there is product market fit, where there is a large team and everything is well-established and understood. Why would you want to visit this research lab with so little merchandise? My dad deserves a lot of credit because he was the one who advised me to accept the position.
Why Was He So Passionate About That?
I had no idea that my dad, a mathematician, was following LLMs in AI and OpenAI until I contacted him to tell him I received a job offer here. He also told me two years ago, something I’ll never forget because it was so prophetic: “Your daughter will tell her grandkids that her mom worked at OpenAI.”
That is absurd! Yes, I am aware. I’ll give him credit because I said, “Dad, you knew more than all my venture capitalist friends.” He’s been my greatest supporter and an inspiration to me.
I was unaware that you had contacted OpenAI. What was it that piqued your attention, according to your friend?
One of OpenAI’s first major clients, Jasper, was presided over by the CRO of WalkMe, the firm I had worked for. He was trying to get me to manage sales for Jasper. When I came across GPT-3 in the documentation when looking over it, I kind of followed a rabbit hole I discovered on LinkedIn. I decided to ping a few folks I knew at OpenAI to find out their opinions of Jasper. One of them then said, “Come join the platform, don’t join the application.” Are you guys creating a sales organization, I asked? He responded, “I believe so.” So, in a hilarious turn of events, “let me introduce you to Sam.”
Did You Foresee The Potential Future Releases Of OpenAI Products?
That felt like the most apparent use case we could figure out, as I believe businesses have been attempting to solve chatbots for forever. DALL-E was there as well when I had my interview. I wasn’t really sure how you would pitch that to an organization at the time. However, once you see something, it seems impossible to ignore. When I first witnessed the power of DALL-E and GPT-3, I instantly thought that everything would change. Now that I’ve seen the technology of the future, I can’t just go work for a traditional corporation.
While I wouldn’t hold it against myself to claim I saw everything coming, I will admit that the models’ enchantment had me completely enthralled. During the interview process, I was instructed to use Python to create an application. The organization is highly technical, and they wanted me to fully understand the capabilities of the model and API. I created a small bot that functioned as a translation application. For me, it was simply utterly wonderful.
Did You Think Sam Had Sound Judgment When It Came To Releasing OpenAI’s Models?
Sincerely, I wasn’t confident when I joined that we would identify the ideal product for the market. I reasoned that this may be a really interesting experiment in which I dabble in AI for a year, learn a lot about it, and then move on to establish a more conventional SaaS startup. At the time, I did believe that the reason it felt so revolutionary was because. It had a very mystical feeling. I simply reasoned that this must have some kind of business use. And I was determined to find out just what it might be.
When Did You First Realize ChatGPT Was The Genuine Deal?
Because we had been experimenting with such models internally for such a long time, I believe we all overestimated the effect it would have. What was so thrilling for us at that moment was that we were all simply focused on GPT-4. An additional source of information and comments on GPT-3.5 was ChatGPT. To be honest, it was a side project; none of us was really concentrating on it. However, after we released it, we soon discovered that it would have a significant impact for which we had not budgeted.
One aspect of OpenAI that I’ve found difficult to grasp is its dual purpose. Building AGI for the benefit of humanity as a whole is the primary goal. The product side has a different tone because it focuses on present, particular use cases.
You are heard. Our sales crew is rather distinct. We are not subject to commission or quotas, which I know surprises a lot of people. Our goals are basically similar: to widely disseminate the advantages of safe AGI. This means that the go-to-market team truly views itself as the AGI sherpas; in fact, we use an emoji to represent this. Our goal is to assist clients and users in making the leap to the AGI paradigm. We do care about revenue, and generating revenue is our main objective. But that isn’t our primary objective. In order to improve our research and models, we also hope to involve our customers in this process and solicit their comments.
Enhancing their internal procedures is one. That goes beyond merely increasing worker productivity; rather, it involves a radical reevaluation of how we carry out our duties and sort of acting as the intellectual foundation for creativity, invention, and teamwork. The second is assisting businesses in creating excellent goods for their end consumers, be it a new shopping experience, a new support system to assist customers in finding answers, or something else entirely.
However, my question is: How does it relate to OpenAI’s overarching, long-term goals? I find the AGI mission to be highly futuristic and long-term.
Yes, you are correct—AGI won’t happen tomorrow. However, there are still steps in the process of obtaining AGI where, with the instruments at our disposal, we can significantly impact businesses and actually turbocharge innovation. Moderna, one of our customers, is one example. They made all of their staff members aware of our ChatGPT enterprise offerings. Moderna is specifically using our sophisticated data analysis tool to assist them in doing analysis on dose data, which they have discovered reduces the average time it takes for them to receive drug approval by thirty days. As they put it, 30 days can mean the difference between life and death for a youngster waiting for life-saving medical attention. Therefore, I see the Moderna narrative as a step toward artificial intelligence. It, quite simply, makes the world a better place by enhancing the ways in which we all develop and create.
Instead of saying to Moderna, “You guys can do this and that in 50 years when we have AGI,” you’re approaching it piecemeal.
The sherpa emoji is useful in this situation. Making progress on this road is more beneficial than having AGI appear out of nowhere. Learning about AI and using it into your workflows and procedures is beneficial. In a similar vein, we find that hearing about how businesses and consumers are utilizing it is quite beneficial. AGI is defined as autonomous systems that are capable of carrying out tasks much like humans. We must comprehend human nature. I believe that it works both ways. It’s beneficial for us to comprehend customer comments and use it in our aim to create AGI. After that, it makes sense for businesses and our users to start incorporating AI into their daily operations.
We are hiring at a breakneck pace while maintaining a strict criterion for talent. We seek individuals who are not coin-operated, but who have a greater stake in our clients’ success, are enthralled with artificial intelligence, are enthusiastic about its potential advantages, and are more cautious when faced with dangers. We are seeking individuals that are genuinely motivated by our aim to fill every go-to-market role.
When you meet with prospective customers, how does the sales team handle the current OpenAI news? For instance, there were problems with ChatGPT going crazy as recently as last week.
What you’re referring to is unknown to me. Examples of ChatGPT users speaking in Spanglish and responding with absurd responses were being posted.
I’m not knowledgeable about that. There’s a lot to keep up with on this site. Really, I don’t know enough about that specific problem.
Could you elaborate on your future goals for 2024?
This year, in my opinion, is really all about making the transition from trial to implementation—which will not be easy. I believe that a well-thought-out plan is necessary. We must be able to assist our clients in embracing change. Breaking away from accepted conventions and more conventional methods can be challenging. This is where the concept of a sherpa will be quite useful.
In a similar vein, it involves transitioning from a sales organization that typically works with startups and tech companies to one that supports Fortune 500 companies. Because we’re dealing with more established industries, we’re also developing as a company, creating resources like an advisory board for customers and event marketing, and we’re becoming quite familiar with the requirements of big businesses.
Organizations are reportedly adopting some generative AI use cases slowly, according to various studies I’ve read. Do you think that would change in the sense of going from the lowest-risk, most internal use cases to something that would actually be implemented for customers?
I completely concur with you on that point. Although there were a lot of options, I believe that at first it took businesses some time to truly grasp how to use this. The decisions that lay ahead practically paralyzed many executive teams and boardrooms. I believe that we’re beginning to overcome it, and businesses are beginning to try new things, experiment, and go into production.
Coca-Cola is an intriguing example, in my opinion. To construct this kind of AI platform where customers can come and use the creative materials from Coca-Cola, they essentially use GPT-4 and DALL-E 3. I thought it was simply such a brilliant approach for a business to use AI to give their customers a whole different experience.
How Is Collaborating With Brad Lightcap, The COO?
Brad is a person who truly has faith in those around him and delegated management of their franchise to them. He doesn’t micromanage. He welcomes challenges, thoughts, and suggestions with great openness. Working with him has therefore been incredibly enjoyable and fulfilling. Another thing I think I truly like about Brad is that he doesn’t have a fixed, unchangeable opinion. He is an adaptable thinker who constantly says to try anything, test it out, and see what works before coming back to me and we can discuss the next steps. That is, I believe, another aspect of the OpenAI ethos that I truly adore: you almost never hear no. If you have an idea and want to do something about it, you’re normally told to put together a proposal, conduct an experiment, and if it’s successful, move on. That is deeply ingrained in the organization’s culture as a whole.
As A Sales Company, Have You Conducted Any Experiments?
All the time! Infinitely! As an illustration, we established a new position in July called Account Associate. I had no idea what this function would entail, but we have faced numerous incoming inquiries and have found it to be quite difficult to reply to them. Rather than relying solely on a standard business development representative to qualify sales leads, I wanted to create a team that could also respond to technical queries from clients. This meant that the team wouldn’t necessarily be a BDR team.
We currently have ten members of the team, and that number will shortly rise to twenty. It’s the front lines, so to speak; we produced an amazing experience. When you contact us for the first time, you will be able to speak with someone who is familiar with our offerings and can help point you in the right direction. I’m not sure if that kind of special job is found anywhere else.
Do You Have Faith That Your Sherpas In Sales At OpenAI Will Climb That Mountain?
We really have a steep mountain ahead of us. I’m not suggesting that everything will be simple. We need to hire a lot of people. These spaces move quickly, therefore we need to keep up with it and be aware of all the various uses. We must guide our clients through it. To successfully implement AI, they must undergo a cultural shift, which we must assist them with. But I have a lot of optimism and excitement. Given the progress we’ve had with a few of our clients, I’m really hopeful that other businesses will adopt this strategy and have comparable results.