Brad Keywell


AI For Good...Maybe a Little Hope?

As someone who lives on the front lines of AI at Uptake, I hear those terms and I react with optimism. Here’s why.

Artificial intelligence (“AI”) and Machine learning (“ML”) are being used to do good. You might not read about it in the headlines, but it’s happening.

They are delivering good not just in the world of capitalism, but also in the world. Social good. Environmental good.

A member of Uptake’s data science team was on Capitol Hill last week briefing policymakers and members of Congress, at an event aptly titled AI for Good. Uptake collaborates closely with Carnegie Mellon University, and together we explained the many ways these buzz-wordy concepts provide more effective and efficient pathways to goodness.

Since Uptake’s beginnings, we’ve acted on our belief that technology companies have an opportunity (and, some might say, a responsibility) to harness their core genius to solve problems that affect everyone. Rooted in this conviction, when Uptake was less than two years old we founded our philanthropic and civic arm --

Our proposition is that technology companies should find areas where their capabilities and expertise can uniquely solve problems that affect society. In other words, we don’t base’s work on writing checks or volunteering. Instead, we believe our greatest philanthropic impact comes from our ability to work directly with nonprofits and the public sector to actually deploy Uptake’s technology and create better outcomes.

We are attempting to actually solve problems. applications have been deployed in pursuit of solving critical issues.

Our passion led us to one of the bastions of data science excellence -- the Carnegie Mellon School of Computer Science. With CMU, Uptake created the Machine Learning for Social Good fund, providing opportunities for faculty and students to apply their genius in AI and machine learning to initiatives that benefit the social sector.

Which brings me back to our time on Capitol Hill, where, alongside our friends from CMU, we discussed these concepts:

  • Open data is critical for AI to improve its machine learning capabilities and impact, and also to democratize access and use;
  • AI is only one piece of the system. There will be users and decision-makers who interact with AI-powered systems, and it’s important to involve representatives of these constituencies as a component of a trustworthy ecosystem; and
  • Transparency is essential to ensure that a system's behavior aligns with our overall societal values.

I’m sure you’ve read about the public backlash regarding this new frontier of AI, reflecting fear that government could be using technology to violate human rights or cause harm. The software industry has a responsibility to self-police this possibility. My fear is that this stifles the possibility of AI doing good for society.

Now, more than ever, we need to be looking for ways to work together, more partnerships, more evidence of AI doing good. With CMU and our activities, we are making a strong case. My aspiration is that, sooner than later, the terms ‘Artificial intelligence’ and ‘Machine learning’ are words that inspire hope, not fear.

Share on social: ︎ ︎

© Brad Keywell 2021