OpenAI's GPT Store has set the stage for the rise of custom chatbots, sparking a frenzy of public imagination. Yet, for May Habib, the Co-Founder and CEO of Writer, the real game-changer lies in crafting customizable, enterprise-grade generative AI that solves real business challenges — safely, securely, and across industries.
First, some quick context. May co-founded Writer back in 2020, building their own large language models well before the days of ChatGPT. Today, the platform empowers any knowledge worker — in marketing, HR, sales, product — to effortlessly generate content, uncover insights, automate workflows, and construct their own applications using a no-code interface. The magic lies in Palmyra, Writer’s set of proprietary LLMs that securely connect to and integrate with customers’ business data.
Listen to the episode on our website, Apple Podcasts, Spotify, or anywhere you listen to podcasts.
One great thing I learned: In the rapidly evolving field of generative AI, the key to maintaining a competitive edge and shaping the future lies not in generic models, but in specialized, purpose-built solutions that are finely tuned to specific needs and use cases.
In the world of generative AI, the real power stems from a combination of domain expertise and continuous fine-tuning. Early on, a series of aha moments led May to bet that generic consumer models would swiftly become commodities (side-eye to you, GPT Store!). To maintain a competitive edge, a tight coupling between data context, client use cases, and AI capabilities is imperative.
But aha moments don't spark in a vacuum. May’s conviction to build Palmyra sprouted from her prior experience leading Qordoba, a machine translation and localization startup. She experienced firsthand how Google Translate democratized and made language translation free…but, at the same time, she saw a surge in a market tailored to enterprise needs.
She expects a similar trajectory for generative AI, where customization and industry specificity become the only defensible moat. It’s fascinating to see how enterprise platforms like Writer can seamlessly integrate corporate brand guidelines, security compliance, and more with niche business data sources and use cases – fueling their flywheel.
As May put it: “We know what Google Translate did to localization. Table stakes localization cost went to zero, but the market got so much bigger…The expectation was that generative AI is going to get built into every surface area of the Internet. And consumer-grade, high-quality, multimodal [is] going to be virtually free.. There is going to be a race to the bottom. And the only moat to build is a moat around enterprise specificity and customization.”
Relying on external LLMs would have imposed a ceiling on Writer’s potential to shape the future of work. Build your own models or lose agency in building the future — that echoes loudly as consumer AI captures headlines.
Of course, this approach won’t be a universal fit. But for the founders and innovators at the intersection of AI and enterprise, it could unlock serious potential. Because while ChatGPT may win our hearts and eyeballs (and give us a deluge of AI girlfriends), purpose-built enterprise generative AI just might be the catalyst shaping the future of business.