Not just a chatbot: Three learnings to help you integrate AI and GPT into your business
by Angus Innes ✦ 21 June 2023 ✦ 5 Min Read
Your LinkedIn feed is jammed with people hyping AI. But how are businesses actually using it in their products and services?
Instead of speculating about whether AGI is around the corner, I sat down with a panel of product experts that are already building with AI to better understand when to use AI and the considerations to make when integrating it.
Read on to hear what businesses including Picsart, Amplitude and trumpet had to say about getting started with AI and applying it to your digital products today.
You no longer need a PhD in machine learning
AI has existed in products for a long time; from Siri on your phone to the recommendation engine that helped fill your last online shopping cart.
Data engineers and machine learning scientists have been working with AI since the early 1980s. What’s different now is that AI is no longer just the output of those who trained for years to understand it. As Henry Kirkness, Technical Director at Planes put it: “You no longer need a PhD in machine learning to start using AI. It's become much like using a database. It’s now a product challenge, a design challenge.”
Start thinking about AI as more of an ingredient, rather than a big technical undertaking.
Generative AI and the conversational interfaces it has enabled make it possible for anyone to use AI and get meaningful outputs they can use in their work, Kathryn Webb, Generative AI Director explained.
With the barrier to getting started now lower than ever, Henry encourages us to “start thinking about AI as more of an ingredient, rather than a big technical undertaking.” In fact, your product teams can start strategising how to use it today.
Think user-first and deliver real value
Despite all the hype, the fundamentals of good product strategy have not changed. Customers don’t care about AI, they care about finding the best way to solve the problem they are facing.
So if you’re going to integrate AI, make sure it's because it meaningfully adds to your user’s experience. You don’t want to come in as a company and say ‘everybody’s using AI tools, so let’s rush this’ Kathryn emphasised.
One way to ensure you avoid the hype and deliver real value is to focus on AI experiences that are invisible to the user.
The best AI integrations are the ones that you don't see. The ones that actually improve user experience. The small ones, the ones that create extra delight, or simplicity. That’s where the value really lies.
To deliver value Will Angel, Lead Product Manager at trumpet suggests using AI to deal with the heavy stuff, the stuff the user doesn’t want to do. That could be things like enhancing search or automating laborious inputs.
And when thinking about measuring AI integration, Joe Reeve, Engineering Manager at Amplitude added: “You measure AI output the same way you measure any feature. Not by looking at the code, but by looking at the outcomes. Look at how your users are behaving and responding to the feature.”
However you’re thinking about AI the most important thing you can do is think about value: what would be valuable for the most important people, which is your end user.
Knowledge share and upskill your teams
So how do you make sure that you’re using AI in the right way? Step one is increasing the base level of literacy around AI in your company.
At Picsart they’ve given full company briefings to their employees. Tata Maytesyan, their Head of Product Strategy & Operations explained: “It’s really important to think about internal learning. We arranged talks for the entire company and ran specific training for all the product managers and engineers to learn about AI.”
Amplitude ran a hackathon to get their team up to speed with the latest AI developments, which actually ended up being the launch pad for their new smart assistant. Will also spoke about running company-wide ideation sessions, saying: “We get the whole company together, from Marketing to Engineering and ideate different ways we could fit AI into our business.”
The best way to demystify generative AI is to break it down into smaller concepts so that you can start applying it to your products, today.
Like with all things new, the best way to understand is to learn by doing. When it comes to the priority skills needed to experiment with AI, prompt engineering stood out.
Prompt engineering isn’t just a fad but a fundamental tool for interfacing with large language models. Breaking AI down into smaller concepts, like prompt engineering will help you and your team understand how you can start applying it to your products today.
Ready to get started with generative AI? Planes are running free half-day workshops for your business to see how generative AI can improve your product. Register your interest.
