What OpenAI’s Plugins mean for Product Managers
Why you need to start working on your own ChatGPT plugin — like yesterday
OpenAI is releasing new updates and functionalities at a dizzying rate, leaving many PMs scrambling to catch up with inefficient tech processes and delayed release cycles.
If you’re a PM reading this, there’s a solid chance you have a roadmap ready for the next 2 quarters. Okay, now I want you to take a look at that roadmap.
Is artifical intelligence a central theme in your roadmap?
No?
Okay, burn it.
You heard me. Burn it. You need to start from scratch.
AI has quite literally changed the tech landscape, making waves like Meta wishes it had with the Metaverse. AI advancements are not stopping any time soon, and B2B companies in particular are at risk because of it.
Time-to-market is more critical than ever. Think about it, ChatGPT has 100 million+ DAUs. Most working professionals (tech and non-tech) are using it in one way or another to automate their work. As more Intelligent systems roll out, while they may not replace jobs altogether, they sure as hell will give an advantage to those who know how to train AI and engineer effective prompts.
OpenAI’s latest announcement of Plugins is a glimpse into a tough upcoming market for digital products.
What are ChatGPT Plugins?
Plugins are essentially add-ons to ChatGPT to access real-time data from specific databases and translate them into natural language out. Let’s take the example of KAYAK’s plugin with ChatGPT. You can basically ask ChatGPT to find the perfect flight for you and book it right there and then. No need to scroll endlessly on KAYAK and compare, ChatGPT will do it all for you.
The first batch of plugins are integrated with the following products/services: Expedia, FiscalNote, Instacart, KAYAK, Klarna, Milo, OpenTable, Shopify, Slack, Speak, Wolfram, and Zapier.
Developers and users alike can sign up to be part of the waitlist and get started using these new plugins. But, what does this mean for product managers?
It means that if we want to remain competitive within the global market, a lot of us need to start thinking of ways in which our products can be directly plugged into ChatGPT, and what that will mean for our data privacy, our revenue models, and most importantly, our product vision.
A few questions to get started thinking about this:
- What points of the customer journey are we aiming to solve?
- Will our solution remain profitable if used within ChatGPT?
- How can we comply with data privacy guidelines and GDPR?
- If our system becomes a ChatGPT plugin, how will we differentiate ourselves from competitors?
What will the future of ChatGPT Plugins look like?
In my opinion, ChatGPT Plugins can go one of two ways. On one hand, they can be curated to serve one use-case per plugin. Competing plugins would need to have significantly different value propositions to exist on the same platform. This would make the User Experience seamless and maintain the illusion of being a single source of truth that ChatGPT gained popularity for. Competitors would create plugins with Google Bard or other competing NLP AIs at the time.
On the other hand, it can become a fish market of Plugins, with every company and their dog having a plugin available. This would take away from the core simplicity of ChatGPT, but make it an extremely powerful single-point tool. Google would need to follow suit, and most companies would have plugins available with both Bard and ChatGPT.
This may be a myopic view of what’s to come, but the truth definitely lies somewhere in between both these extremes.
What do I, as a Product Leader, do next?
Revisit your customer journey map/business model canvas/whatever you use to define your product’s value to your end customer.
Find ways in which you can integrate AI within core product flows, such as automated form filling, suggested information, insights on company-level product usage and performance, chatbots to automate key flows.
Now think inversely: what can you offer OpenAI so that it plugs your services to its users? The most valuable information companies can have now is data within context: of users, of user behaviour, of historical events within context.
So, get started. Create a long-term vision for your product in both cases, whether it is powered by AI or it is empowering AI.
Sign up for the ChatGPT plugins waitlist.
Explore upcoming open-source models and find ways to collaborate with them ASAP.
By embracing AI and ChatGPT Plugins, you’ll be better equipped to navigate the rapidly evolving tech landscape and ensure your product remains relevant and competitive in the years to come. Don’t wait — the future of your product depends on how quick you are to adapt and innovate in this new, crazy, AI-driven world.
I write frequently and primarily about product management, mainly tech products within B2B spaces. If you want to see more content like this, consider following me. Thanks!
