By John Q. Hosedrinker 
The United States Postal Service maintains a comprehensive system for addressing mail, using a standardized format to ensure efficient delivery. This system relies on a two-letter abbreviation for each state, territory, and associated region, such as “CA” for California. Additionally, postal codes, or ZIP codes, are crucial for sorting mail, directing it to specific geographic areas for final distribution. This comprehensive system is essential for organizing and ensuring timely delivery of mail across the country and to international locations.
Read the original article here
U.S. court rejects Toronto AI startup’s bid to dismiss copyright infringement suit. This news, quite frankly, is not surprising. The legal battles surrounding AI and copyright are just getting started, and this is just another domino to fall. A Toronto-based AI startup, now facing a copyright infringement suit in a U.S. court, has had its motion to dismiss rejected. That means the court believes there’s enough merit in the allegations for the case to proceed.
How are AI companies supposed to make a profit? That’s the million-dollar question, isn’t it? The sentiment I’ve picked up, and quite frankly, understand, is that the current approach feels a bit parasitic. The idea of profiting from the unauthorized use of others’ work rubs a lot of people the wrong way. The suggestion is, they’re essentially stealing content, and the courts seem to agree, at least in this instance.
Why is a US court ruling on what a Canadian company is doing? Well, the fact that this lawsuit is in the U.S. federal court, specifically the Southern District of New York, is key. This hints at the possibility that the Canadian company is offering its products or services in the U.S. or that their actions are impacting a U.S. entity. If a commercial entity offers their product or services into USA, or injures an American citizen (e.g., corporation) within the US, then that is usually enough for jurisdiction.
The central issue is the alleged copyright infringement. This is the core of the case, and the court’s decision to deny the dismissal means they see a valid claim here. The claim is that the AI company has used copyrighted material without permission. This is especially true because the court is likely to find that the AI company has made use of copyrighted works.
It appears the court may be looking at the case in a similar way to other recent copyright infringement cases involving AI. The fact that the court denied a motion to dismiss means that the allegations, if proven true, constitute a legal wrong that can be addressed by the courts. We’ve seen this before. For example, the case of The New York Times against OpenAI, where a similar motion to dismiss was also denied.
I think the financial model of these companies is precarious. Some say the investment needed to get AI off the ground is huge, and I mean, HUGE, so the return on investment needs to be high to get an acceptable return on investment. The idea of relying on “stolen” data to train these AI models just seems unsustainable.
And the question of how AI companies make money is also interesting. There’s licensing their models to other companies, which could be part of the revenue stream. Others use the data and try to sell your information to advertisers, which is not exactly comforting.
So, where does this leave the Toronto AI startup? Well, they’re now in for a legal battle. And there might be a lot of them. As the lawsuits pile up, it’s quite possible that we will see a massive market contraction. This means only a few major players might survive.
What happens if AI companies start paying artists? Well, it might be the only ethical way to do it. But it may involve changing the business model. The training of AI on a small data set can generate a decent amount of revenue. The alternative is the use of open-source data. However, the data sets aren’t going to be as interesting, and they aren’t going to have the copyrighted characters in them.
In short, there is a way to do it right. The question is, will they?