Why AI Hackathons Won’t Build Solutions to real-world problems

Why AI Hackathons Won’t Build Solutions to Real-World Problems

Here is what we should do instead to build AI solutions successfully.

There are dozens of hackathons that are being organized these days in response to COVID-19, 1000s of people are giving their time to build solutions. I have mentored one of the largest in Europe, where over 1000 engineers participated. The organizers put days, if not weeks, of work into it. So I truly commend their efforts and goodwill. However, without taking away anything from the organizers and participants, I question the effectiveness of hackathons.

Why hackathons won’t build real-world solutions

1. Lack of domain expertise: Social problems like COVID-19 cannot be solved only by engineers. To build real-world solutions we need to involve policymakers, domain experts, users — something that teams participating in hackathons often don’t have access to.
2. The absence of diverse backgrounds leads to bias in models: Hackathons are formed by teams who know each other and thus end up being of the same background. This misses out other opinions and backgrounds. We have seen that systems developed only by engineers end up being biased. Alexa and Google Home are being used by men to lock out their spouses in instances of domestic violence. They are turning up the music really loud, or they are locking them out of their homes. It is possible that in an environment with mostly male engineers building these products, no one is thinking about these kinds of scenarios. Additionally, there are many instances about how artificial intelligence and data sensors can be biased, sexist and racist [1].
3. Too little development time: Three or four days is not enough to test or build solutions replicable in production environments. Remember that in one you are impressing judges but in the other, you are building entire solutions. Building a prototype is easy, but there are tens of other external entities that need to be considered when moving into the real world. Besides technical challenges, there are other areas of focus that need to be integrated with the prototyping (such as marketing, design, and sales).
4. Lack of diverse ideas . When I was mentoring hackathons, I can say that out of 100+ ideas — we can put all of them in 10–15 ideas. Most of them were similar — and that is because of point 2 mentioned above.
5. Competition vs Collaboration: Why do we need 100s of teams to compete with each other when they can collaborate? In one of my previous articles, “ Why Kaggle Is Not Inclusive and How to Build a More Inclusive Data Science Platform ”, I argue that competition-based models like Kaggle are not the best model to build real-world solutions. Amongmany reasons, a key issue is that people are incentivized to win instead of working together to find the best solution to a problem. The shortcomings of competitive models also apply to hackathons.

The alternative? A Collaborative Approach

If we are serious about solving social challenges like COVID-1 9in an inclusive and effective way we need to ensure several mechanisms.

1. Harness global collaboration : Most of today’s challenges are global in nature. It is not that one country or group of people can solve it. So to solve such challenges, we need a new model, a model where the global community can come together to solve problems, share their data and build solutions. In the world of Big data, AI and Machine Learning, data is the key. It is not a sophisticated algorithm or a better team, but it’s the team with the better (and more) data that wins. In addition, due to online courses, education, especially of emerging technologies like AI and ML, became easily accessible. Now, for anyone in any part of the world, it is very easy to start learning from websites like Udemy, Coursera. These have made it possible to get a community of highly motivated people who have a common vision, motivation, and mission and create global collaboration.
2. Work closely with domain experts. Most real-world problems are not limited to just a data science problem but involve domain experts to create value. We have seen that while working with domain experts, data scientists from diverse backgrounds help the company to refine the problem and give a new perspective to the problem.
3. Involve people who face the problem: By incorporating people from all backgrounds, especially people who faced the problem. This brings empathy and non-bias as people who face problems will inherently not build something that is bad for them.
4. Follow a bottom-up model : I firmly believe the future of innovation is bottom-up, where communities will come together to collaborate and solve their problems. An article I wrote on the same How Communities will drive the future of AI, not Governments or Corporations . I argue that communities have both intrinsic and extrinsic motivation to solve the problem, which is essential in building future solutions. How great it is when people who face the problem, actually are able to build the solutions.

“Top-down collaboration does not work, its time to build a bottom-up model”

In conclusion, I am arguing:

We should think more holistically and do our best to create the right environment where we look beyond gender, race, and cultural background and focus on how we can collaborate as humans to build a better future. Hackathons do not create that environment and thus not best suitable for building real-world solutions.

Digitization and AI have enormous potential for doing good in all aspects of life and in all sectors of the economy. However, it is the combination of people with technology that truly enables progress and higher productivity. We have to emphasize the community and the right people with a purpose. That is actually the key to create future solutions and products.

References:

[1] https://www.logically.co.uk/blog/5-examples-of-biased-ai/