NYT bestselling authors and impact-driven entrepreneurs on the art of building lean AI startups.
This article will share insights on how to build lean startups that change society for the better and leave a positive impact on the planet.
There are hundreds of use cases where AI can help to do exactly this. Impact-driven startups have the potential to solve real-world problems, tackle environmental problems, and improve the lives of many people, especially vulnerable populations.
Billions of dollars are already flowing into AI ventures, which are primarily addressing profit gains and industrial automation. The AI for Good movement where often commercial meets social value is slowly picking up.
Now, in order to build impact-driven AI startups, there are a few essential steps to follow.
As discussed in the bestselling book “The Lean Entrepreneur”, our media glorify business outliers like Bezos, Branson, Gates, and Jobs as heroes with X-ray vision who can look to the future, see clearly what will be, imagine a fully formed product or experience, and then, simply make the vision real.
In reality, to be visionary, one should not merely execute on a seemingly good idea and ignore all doubt. Otherwise, startups will build doomed products in a vacuum where there is no real product-market fit, which leads to costly and heart-breaking failures.
The alternative scenario is to go lean right from the start.
Brant Cooper, NYT bestselling author of the “Lean Entrepreneur” and CEO of Moves the Needle recently spoke at one of our demo days and gave the following definition:
The idea around lean is the elimination of waste. Don’t waste your time, money, resources, creativity, inspiration in building products nobody wants.
1. How to automatically deskew (straighten) a text image using OpenCV
2. Explanation of YOLO V4 a one stage detector
3. 5 Best Artificial Intelligence Online Courses for Beginners in 2020
4. A Non Mathematical guide to the mathematics behind Machine Learning
You have figured out a problem, now it’s time to start testing. The good thing is that there is lots of open source code, which won´t cost a dollar and can be used to build an MVP. This will also help to test if you really need AI at this point in time.
Just because AI is an approach to solving your problem, it might not be the only way to do so. Each case is different, but at this point, when you’re using real humans or some other way to test your problem, stop and ask yourself the following question:
What is it about my problem that AI could solve in a better (more accurate, less costly, more reliable, etc.) way?
If AI proves to be a promising direction, you´ll need to gather some data, curate it to make it’s useful, then design a model, and train it.
In practice, the work required to find, curate, and manage the data is often the biggest and hardest part of the problem.
However, there is also a common misconception that you need a perfect data set to get things started. This is not only technically wrong but also would mean that the vast majority of organizations will never be able to do any meaningful work with AI. Often, a promising direction is to leverage the growing amount of open-sourced datasets.
Making progress in AI means to take incremental steps towards creating higher returns on impact and value rather than looking for the costly “perfect shot”.
Now it is time to package your AI up into a product with a user interface and other features than just doing AI.
Remember, a good product solves a real-world problem. It’s no good having an AI that can just look at a photo and differentiate between cats and dogs.
Today, still most AI is developed in isolation from the people and social circumstances that make them necessary in the first place. To change this, we need to enable truly human-centered development by moving away from old ancient practices and top-down approaches.
A bottom-up approach takes the development process out of the lab and into the real world. This embeds diverse perspectives and open dialogue, where the user or customer of the solutions is part of the development process. All of which increases the chances to build inclusive and trustworthy solutions.
The good thing is, if you are solving a real-world problem, people are more likely to join your endeavor and give feedback.
The better the data you can give an AI when you train it, the better it is.
Once you’ve launched your startup, you’re going to start gathering more data. Data that you didn’t have when you first trained your AI. Now, you can improve your AI. In order to do this effectively and efficiently, you need to raise a few detailed-oriented questions in your team early one, such as:
- How are you going to gather and store data?
- How often and when are you going to retrain your AI?
- How are you going to test performance improvements?
Lastly, I want to share two very inspiring real-world examples of applying AI for the betterment of society.
Detecting child malnutrition through computer vision
Child Growth Monitor is a game-changer startup that is working on an application to identify malnourished children under the age of 5 years.
It is a solution based on a mobile app using augmented reality in combination with artificial intelligence. By determining weight and height through a 3D scan of children, the app can instantly detect malnutrition.
Malnutrition in children is a problem affecting more than 200 million children.
Anticipating shocks to deliver life-saving health products
VIEBEG is building AI to anticipate shocks of all kinds to deliver life-saving health products in challenging environments. VIEBEG is a leading provider of medical products and equipment and the first online pharmaceutical distributor in East Africa, providing more than 300 hospitals and clinics in Burundi, Rwanda, and DR Congo.