Algorithms and AI can be helpful in the fight against racism, but they can also be harmful; it depends on how we use them. Think of it like a kitchen knife – it can help prepare a healthy meal or be used as a weapon.

One positive way AI can address racism is similar to how Doordash works. Their app assigns deliveries based on how close the driver is, not on their name or background. In the same way, AI can be used in job applications to hide information that might lead to unfair bias when selecting candidates for interviews. This focuses the decision on skills and experience, not race.

AI can also fight racism by finding and removing hateful comments online. Algorithms can scan social media for racist language, allowing companies to take it down and create better online communities. Another positive way to use AI is in loans or credit checks, where algorithms can consider more factors than old systems did, potentially helping minority applicants get fairer treatment.

But there's a negative side. If AI is programmed with information that is already biased, it will just keep making unfair choices. For example, facial recognition software can easily become inaccurate and racially biased, unfairly targeting people of color. Also, the Doordash app, while good for drivers, might still result in fewer deliveries for those in poorer neighborhoods where people don't tip as well.

To make AI a helpful force against racism, we need to provide it with diverse information to learn from. We also have to have humans double-checking its decisions and regularly checking the algorithms themselves to make sure they aren’t accidentally making things worse for minorities.