No. 217: Algorithmic Self-Hate

There’s been a lot of conversation around the importance of making sure there’s diversity among developers and researchers who are building artificial intelligence. One of the primary concerns is that this diversity imbalance will lead to bias being built into algorithms. The proposed remedy is for teams to hire more minorities on their teams.

I’m a strong proponent for diversifying the makeup of those building artificial intelligence. But, what does one do if the minorities one hired to join a team continue proliferating bias in their code, bias that hurts them?

I recently heard a story about a group of minority developers building AI algorithms that wouldn’t have recognized their skin color. I don’t think this is something that has gotten much of any attention, but it’s something worth examining. Just because someone is a minority doesn’t mean they haven’t digested tropes about white people and those of color. Are these folks in the right mental space to think independently and proudly? Do they see color or pretend not to?

I’m going to be pondering this more, but wanted to put the thought out there. What frameworks do have in your backpocket for thinking through issues like this?

No. 216: Migration in Africa | International AI Principles | Cut Through the Hype

Op-Ed: The Future of Africa’s diaspora is in Africa (CNBC)

42 Countries Agree to International Principles for Artificial Intelligence (NextGov)

Five questions you can use to cut through AI hype (MIT Technology Review)

No. 215: Congressional vs. Asia Pacific AI Investment | AI with a Memory

Heinrich, Portman, Schatz Propose National Strategy For Artificial Intelligence; Call For $2.2 Billion Investment In Education, Research & Development (Martin Heinrich)

I wrote here about Congressional efforts to figure out the U.S. artificial intelligence direction. That effort plus this bill by a group of senators helps move us closer to the U.S. having an AI strategy. That said, $2.2B is not nearly enough investment. Just consider salaries for AI researchers – Google’s Deepmind lab paid $138M in salaries to 400 employees back in 2016.

IDC: Asia-Pacific spending on AI systems will reach $5.5 billion this year, up 80% from 2018 (TechCrunch)

This shows what the US is up against when it comes to investing in AI research and development.

Artificial intelligence becomes life-long learner with new framework (Science Daily)

As the son of a Wolfpack alumna, I had to include this. Further, as one with a long memory, I don’t know how I feel about AI software getting back at remembering previous tasks.

No. 214: Monday/Tuesday Reads

Physics-based AI software company joins Cottonwood Venture Partners’ leading digital oilfield portfolio (Tachyus)

Sam Altman’s leap of faith (TechCrunch)

Sorenson Seeds $150 Million Fund to Invest in Overlooked America (Bloomberg)

No. 213: Is That the Whole Story?

I’m working through The Quest for Artificial Intelligence by the late Nils Nilsson, a pioneer in the development of the field. He charts out the foundation for artificial intelligence beginning with Aristotle’s syllogisms and makes his way through various European mathematicians who made various contributions to mechanizing logic.

A couple of years ago, Chris Dixon, a venture capitalist at Andreessen Horowitz, wrote a piece on how Aristotle laid the foundation for the creation of the computer. Is that the whole story? I think there’s more to it than that. I tweeted about this early last year.

It’s critical that black folks and other groups currently underrepresented in the development of artificial intelligence carve out a space for ourselves. I wrote a bit about why here.

Perhaps a good example of the impact carving out a space can have is in the news that Robert Smith committed to pay off the loans for the entire 2019 class of Morehouse graduates.

Smith has built Vista Equity Partners into a machine of a private equity firm that has outpaced its competitors investing in enterprise software businesses by executing a precise operations playbook in each of its portfolio companies before flipping them for real nice returns. Pitchbook estimates that as of 2017, Vista’s internal rate of return has averaged 22%. Compare that an industry average of nearly 10%, according to AQR Capital Management research. Smith has done quite nicely for himself as a result, generating the resources to be able to clean up $40 million in debt.

A key engine behind the playbook Vista Equity Partners deploys across its portfolio companies is Vista Intelligence Group. The group uses artificial intelligence to scan data across Vista’s portfolio companies to surface opportunities and stand up new businesses around them. In a fireside chat at Goldman Sachs, Smith talked about how Vista is navigating the fourth industrial revolution by trying to get the firm to a leadership position in the various ecosystems it invests in rather than just placing capital in particular narrow verticals. I venture that Vista Intelligence Group is the lever they’re turning to make that happen. Artificial intelligence is the magic sauce. Here’s that fireside chat.

Imagine it’s true that artificial intelligence is the new electricity as Andrew Ng claims. Now consider that Robert Smith has leveraged AI to generate $4.47B worth of resources personally while managing $46B. Imagine the possibilities of some of the 2019 Morehouse graduates going on to tap into their genius and reimagine AI and how it will shape the global economy.

What do these graduates need to tap into their genius? The lightened financial burden courtesy of Mr. Smith definitely helps. Another component is these graduates seeing beyond the narrative that the arc of technological development cuts through Europe. I’ve written about this narrative issue here. While Aristotle was developing syllogisms, equally brilliant philosophers and mathematicians were working on their own ideas across Africa. Drawing confidence and inspiration from that kind of foundation makes these graduates unstoppable in my mind and positioned to reshape the trajectory of this world.

No. 212: Track Saturday

The Shanghai Diamond League meet is this weekend, and the 100m dash was electric.

Christian Coleman is an incredible sprinter. He holds the world record in the 60m dash. Over the first 50m of this race, you see how. He’s like a loaded spring out of the blocks. He loses this race in similar fashion to the way in which he lost races to Ronnie Baker last year – he loses steam towards the end of the race. I think the biggest reason why is that his body remains in a quasi-drive phase throughout the race. That’s putting extra pressure on his body to remain balanced while also trying to stave off deceleration. All-in-all, it’s inefficient and his body gets tired quick.

What Noah Lyles does from the middle of the track in this race is not normal. A professional sprinter should not run down other professional sprinters like this to win a race, yet he does in large part because of how efficient he is after he gets out of his drive phase. Watching him run is a thing of beauty. His start is quite weak, but considering he’s three years out of high school, he’ll figure it out. Once he does, he’ll come the closest to breaking Usain Bolt’s 200m dash record and just might break it because he actually runs well through the finish line.

No. 211: Exposing disadvantaged folks to AI

One of the things that keeps me up at night is the thought of underrepresented minorities and poor people not being at the table as the latest AI developments emerge. So, I was through the moon when I listened to this interview with Tara Chklovski, founder of Iridescent, a nonprofit committed to exposing disadvantaged families to AI.

Check out the interview to learn more about their work.