Dear Reader,
Welcome to the weekly mailbag edition of The Bleeding Edge. If you have a question you’d like me to answer in a future edition, write to me by clicking right here. Let’s get started…
Hi Colin: First, I enjoy the newsletter and your contributions to the articles. I was wondering if anyone has considered AI for use in the courtroom because I know there are judges biased this way or that way, and jurors also have their bias toward one thing or another. I’ve talked to people who have been on jury duty and admitted that they changed their vote because of pressure and also to get out so they could attend their child’s game.
I’ve always thought how terrible it’d be to send somebody to jail because you wanted to see a game… If an AI were used, it could be loaded with the Constitution and the Bill of Rights, plus the local laws of the state, etc. Then, all of the testimony from the defense and the prosecution could be fed in.
Then, the AI could go through all the data and come up with an unbiased conclusion and a fair punishment. That way a person can be treated and sentenced in a fair and just way with a punishment that fits the crime. Looking forward to your comments.
– Jim V.
Jim, thanks for the message and sending in your thoughts. I believe AI becoming the “arbiter of truth” would have a profound impact on society. Similar to how a calculator is the arbiter of math, I can see a day where AI has that ability for a wide range of use cases.
The technology is close. In many ways, we’re already at the point where AI could be the final ruler in a court case. The more complex cases that involve photographs, video, and elaborate testimony are probably safe from AI taking over as there’s only so much a system can achieve.
However, cases that involve breach of contract, arbitration, or are civil in nature could be litigated with AI as the decision maker. I could see a scenario where the two parties agree to use AI to speed the process up and/or reduce costs.
Probably the highest hurdle isn’t the technology, though. It’s getting the public to accept the AI decision as if it were a human judge. It’s difficult for me to pinpoint when that time will come, but in many ways it’s inevitable. The court system is overcrowded and slow. And also as you state, maybe more importantly, AI could offer more fair rulings.
I could also see the topic of an AI as judge and jury being a contentious issue for constitutional scholars. But, interestingly, the literal text of the sixth amendment guarantees the right to a “speedy and public trial, by an impartial jury.”
I could see an argument being made that an AI jury would be speedier and more impartial than the current system. Only time will tell if the legal system ever accepts that in the future.
The problem I see with AI is “garbage in, garbage out.”
– Steven T.
Steven, thanks for the comment, and it’s worth discussing.
You’re correct, an AI system is only as good as the data it’s provided. There are obviously other factors involved – but gain access to the highest quality Nvidia GPU chips with garbage data and you wouldn’t create anything very useful.
The fact “garbage” data exists is true about the internet in general, so I don’t view this as a new phenomenon exclusive to AI. It’s truly been a problem since the internet began. When we log onto sites like Twitter or Reddit – we expect to see some “garbage” or certainly opinions that aren’t necessarily backed up with any real data or facts. In many ways, the sites are designed to promote this type of content since it draws strong engagement.
Ultimately, users posting large amounts of meaningless or garbage data isn’t going to allow an AI company to win. Quality data is going to determine who wins or loses. We’re already seeing the lines drawn, and it’s heating up.
Last week, The New York Times changed its terms of service to prohibit using its articles in AI training. Websites are now adding code in an attempt to block website crawlers from scraping data. We’re going to see more paywalls, sign-ins required to view content, and the internet is going to become less open and free. That’s honestly going to be an unfortunate byproduct of AI data being so valuable.
The truth is, as much garbage data that is out there, there’s a treasure trove of good data – and its value only increases with AI. As we’ve highlighted with a few of our stock recommendations, the companies that have direct access to high-quality data will gain a huge advantage. Consequently, the companies that don’t have access to it will fall behind.
With the hype and projections surrounding AI, I don’t hear much about where all the power will come from. EVs, robots, blockchains, and supercomputers require a constant supply of juice. Factories with robots can run 24hrs a day all year long, which means the machines they operate will run all day as well. That’s a 4X drain right off the top.
– Roger T.
Roger, thanks for writing in. This is a great observation and is absolutely forward-thinking.
In many ways, the country that solves the energy challenge wins the AI race. I’m not confident that the federal government can agree and will move fast enough in this area – so the innovation will likely fall to the private sector.
Right now, there are some news stories about the energy (and water) usage to operate these large data centers. I believe the coverage will increase as we move into 2024.
We haven’t experienced it yet, but at some point, we will have an event or a series of news articles that force the data centers to be more proactive about power and water management. Nike and Apple have experienced this as it relates to poor working conditions in manufacturing facilities – we’ll have this moment as it relates to data centers soon.
Luckily, these large companies are fairly flush with cash, more so than the federal government in some ways, so with enough public pressure the innovation will occur.
We’re currently looking at several companies to recommend in this area. There are technologies not being used in large data centers that will both reduce energy consumption and water usage. I look forward to bringing you those recommendations very soon.
That’s all for this week’s mailbag. Thank you to everyone who wrote in.
If you have a question you’d like me to answer in a future mailbag, feel free to send me a message at [email protected].
Regards,
Colin Tedards
Editor, The Bleeding Edge