‘Seasoned’ advice from Pepper the robot

Sabu Samarnath
Sabu Samarnath
3 min read

UPDATE 14/08/2017: Pepper’s cocktail integration has also featured on TVB’s Money Magazine, a leading Hong Kong business news programme – watch the video here.

Together with support from WinWin – a partner of SoftBank, the company behind Pepper – Rainbird has created and integrated a decision-making engine which enables Pepper to interact with users, asking questions about their flavour preferences, driving habits (as Pepper says, “Always drink responsibly!”), and how exactly they feel about a robot serving them a cocktail. At the summit, Rainbird’s intelligent platform combined these responses with real-time data sources, to recommend cocktails tailored to delegates of the AI Summit. Who says you can’t have a tipple before lunchtime? Not Pepper, that’s for sure, as the friendly robot gave the go-ahead for numerous Bloody Marys…

Making the tough decisions (like which cocktail to choose…)

Although it is serious fun, Pepper and Rainbird’s light-hearted integration is more than just a gimmick. The Rainbird platform, in combination with Pepper’s engaging and sensitive interface, demonstrates the flexibility and capability of cognitive reasoning, a form of AI which is analogous to the human decision-making process. Unlike machine learning, which requires large amounts of historic data upon which to base a judgement, cognitive reasoning can deliver accurate, traceable decisions created using expert knowledge, combined with numerous data sources, and information gleaned from its questions. The intelligent reasoning behind Rainbird is demonstrated by the platform’s ability to make nuanced, complex decisions and create responses based on numerous variables. In the case of the Pepper integration, this was limited to: whether the user was driving later; their individual mood (assessed through Pepper’s interface, using IBM Watson’s sentiment analysis); their taste preferences; and the time of day. In a business context, a similar interaction could just as easily solve a customer service problem, or offer advice on a tailored financial services product.

There is practically no limit to how many human or data-driven factors Rainbird can take into account when making recommendations, performing diagnostics, or solving problems. Since Rainbird is independent of the user interface, its single system can be delivered through any open interfaces, including: Amazon Echo, Google Home, Facebook, SMS, or even all of these concurrently.

Beyond Pepper

Yet, the enquiry and decision-making demonstrated by Pepper’s cocktail advice is just the tip of the iceberg for what Rainbird can do. Rainbird’s proven AI solutions highlight what it can achieve for enterprises when working with larger knowledge models and data sets. One key innovation of the Rainbird platform is its ability to work in a non-linear way – handling numerous lines of questioning, and commencing user interactions from any point on its ‘knowledge map’. It can even provide answers in the face of missing information or uncertainty in its model, by soliciting important data from its user.

Rainbird can also discover new facts, taking previous outcomes and applying them in its future interactions. If we consider this in terms of our cocktail advice demonstration, Rainbird can remember your previous cocktail preference (a Mojito, please!) and increase the weighting towards this kind of drink in its subsequent interactions with you, or even people like you.

Whilst Rainbird’s latest demonstration was undoubtedly a bit of fun, the scope for enterprise solutions, using the intelligent decision-making of the Rainbird platform, is abundantly clear. But, for now, when you just can’t decide between a Cosmo and a Corpse Reviver, what more could you need but a friendly Pepper robot to show you way?

Watch Pepper hard at work delivering Rainbird’s intelligent decisions

Transform your business into a Decision Intelligence powerhouse

Explore how Rainbird can seamlessly integrate human expertise into every decision-making process. Embrace the future of Decision Intelligence powered by explainable AI.