I’ve been quite open about the many ways AI-driven projects can go wrong. With so many ways to fail, what makes Sumo Logic Mo Copilot set up for success?

1. Strong use case

2. Clear vision

3. Dedicated user experience

4. Continuous feedback and improvement

Let’s discuss each of these points in detail.

Strong use case

One of the most significant failings of many AI-driven projects is the lack of a clear use case where AI can directly tangibly benefit a customer. This could not be further from the truth with Copilot – the user benefit is clear and immediate.

Anyone who has worked with Sumo Logic knows it’s the world’s most powerful and performant log search engine. They also immediately recognize that you need to know a lot of Sumo Logic Query Language syntax to unlock this power and performance. As Robert Sheckley famously said, “To ask the right question, you need to know part of the answer.” Historically, knowing how to ask the right question has always been a challenge for new users of Sumo Logic or those who need the power of logs only occasionally.

With Mo Copilot, “knowing part of the answer” is no longer an issue – the entire experience is set up around using natural language to query the world’s most capable logs database. You simply tell Mo what you are looking for, and it does the rest, quickly converting your question into the Query Language, running the query for you, and displaying the answer.

From a design perspective, it’s particularly exciting to see how Copilot can create visualizations based on your data. Even power users can sometimes struggle to build the best dashboards in any tool, but with Mo Copilot, powerful queries such as geographical distributions and complex time series transformations are easy to visualize.

Clear vision

Many AI-driven projects are a proverbial “hammer in search of a nail,” where technology capabilities drive the features and user experience. By contrast, Mo Copilot never lets you leave the Sumo Logic Platform empty-handed. By design you can always get to the next step – deep insights and impressive capabilities are always within reach.

Copilot does this in two key ways: autocomplete andnext steps suggestions. Autocomplete has been around for a very long time, but this is the first time it’s driven by a powerful AI engine. Gone are the days of staring at an empty search bar, wondering where to begin. Mo can recommend initial starting points, query auto-complete, and even source expressions:

Note the arrows to the right of the autocomplete: those let you populate the query in the search box for further editing instead of simply running it, saving both time and money and making maximum use of the autocomplete feature.

The second visionary feature is the Suggestions area on the right side of the screen. Those are not pre-canned suggestions but are highly customized and driven by the user’s journey through the system. Copilot responds to and continuously learns from the user, always striving to present the most insightful exploration ideas leveraging the industry’s deepest knowledge of log search best practices:

Through these design principles, users of all levels of expertise can easily interact with Mo and find the insights that they need in the Sumo Logic Platform.

Dedicated user experience

In partnership with our customers, the user experience team has thoroughly researched, tested, and validated every aspect of the Mo Copilot user experience. When testing various options, it quickly became apparent that rather than taking a cheaper and more common “postage stamp” side panel approach, our customers needed a brand new way to interact with Sumo’s powerful log search engine: a dedicated set of pages on which the Copilot experience could unfold.

Having a dedicated experience allowed us to implement autocomplete and suggestions in the most effective fashion, as well as introduce a powerful restatement feature, where Copilot would echo back to the customer how it interpreted the query and provide an easy way to see the Sumo Logic QL query Copilot created. This allowed customers to validate the accuracy of the Copilot interpretation and continuously learn more of Sumo Logic QL, engendering instant trust and long-term loyalty:

Continuous feedback and improvement

AI products are a different breed of technology because they are not programmed – they are trained. And no training is ever complete.

The continuous improvement process is engineered deeply into the DNA of the Mo Copilot. Not only is every query and click leveraged to the fullest, but the UI includes key features that allow the user to report any and all opportunities for improvement immediately and directly to the system creators:

Even small but important touches like copying the query into the “expected” box gives a way for knowledgable users to provide direct curated input into improving the model.

Final thoughts

Designing an AI product is a balancing act of showcasing the power of AI and also making it feel effortless and easy. If it feels too easy, you might not trust the results. If it appears really powerful, you might find it too challenging to use.

In designing Sumo Logic Mo Copilot, we aimed to give it the breathing room needed for it feel grand and powerful, while also intuitive and helpful. This nuance gives users like you access to what’s possible in the Sumo Logic Platform with a UI that you’d expect in our AI-powered future.

Try it for yourself with a 30-day free trial and see how the UI for this AI capability is built for success.