Member Article

Applause re-invents AI testing with new solution that detects bias and sources training data at scale to make apps more human

Applause, the worldwide leader in digital quality and crowdsourced testing, today announced its new solution for AI training and testing.

The scalable solution trains algorithms to learn quickly and tests the output to ensure those algorithms are processing and responding appropriately.

The solution leverages Applause’s global community of vetted testers to deliver the widest possible range of training inputs. The results are then tested across every possible device, location, and circumstance to identify issues and provide actionable user feedback in real time. This enables today’s leading brands to identify issues of quality or bias earlier in the development process so that they are ultimately delivering top-quality AI experiences for their customers.

“Users want AI to be more natural, more human. Applause’s crowdsourced approach delivers what AI has been missing: a diverse and large collection of real human interactions prior to release,” said Kristin Simonini, VP of Product at Applause. “Not only will this improve AI experiences for consumers everywhere, the breadth of the community also has the potential to mitigate bias concerns and make AI more representative of the real world.”

The data Applause collects from the community comes from people across numerous countries, ages, genders, races, cultures, political affiliations, ideologies, socioeconomic and education levels, and more. This broad base of data samples results in a more representative and unbiased output than if the data were sourced from a smaller group.

Not only does the Applause Community provide diverse training data sets to power algorithms – it can also test the outputs of those algorithms to check for bias. If bias has crept into an algorithm at any stage, the community can identify it when testing the output, something that a smaller or less diverse group of testers might not be able to do.

All types of AI – from virtual assistants learning how different users ask for the same thing to nutrition apps identifying food from uploaded photos – have been hampered by the same challenge: sourcing enough data to teach the machine how to interpret and respond, and then testing the output at scale to ensure the results are accurate and human-like when necessary. Applause’s solution addresses these challenges directly, across every vertical and platform.

Specifically, Applause’s new solution operates across five unique types of AI engagements:

Voice: Source utterances to train voice-enabled devices, and test those devices to ensure they understand and respond accurately.

OCR (Optimized Character Recognition): Provide documents and corresponding text to train algorithms to recognize text, and compare printed docs and the recognized text for accuracy.

Image Recognition: Deliver photos taken of predefined objects and locations, and ensure objects are being recognized and identified correctly.

Biometrics: Source biometric inputs like faces and fingerprints, and test whether those inputs result in an experience that’s easy to use and actually works

Chatbots: Give sample questions and varying intents for chatbots to answer, and interact with chatbots to ensure they understand and respond accurately in a human-like way.

Many AI experiences today fail to meet customers’ expectations. Applause provides rapid, iterative feedback from an end user’s perspective, at all stages of development, to ensure brands are creating an experience customers will want and use.

This was posted in Bdaily's Members' News section by Martin Withers .