The Open AI logo, which represents American research into artificial intelligence (AI) … (+)
There’s usually a lot going on with OpenAI, one of America’s leading artificial intelligence companies – and this week is no exception.
There is a roundtable of related stories and information about the industry, showing different aspects of how this company is contributing to smarter models and greater impact on our world.
The first is a case study that demonstrates how businesses can use OpenAI technology to pivot in a big and immediate way.
Modern Banking with ChatGPT
Tech journalists report that BBVA bank has ordered 3,000 ChatGPT licenses to create AI entities that will help with tasks such as credit approval, account analysis and other administrative formalities.
For reference, OpenAI claims to have (literally) a million enterprise customers, and that’s just one. But you can see the potential for job disruption and business efficiency in the way the bank configures its AI agents to function like human task-based workers. The Wall Street Journal reports how the Madrid-based bank integrated ChatGPT into its risk, marketing, talent and legal departments, with the ChatGPT enterprise tool. As WSJ journalist Isabelle Bousquette writesBBVA is not alone, although questions remain about scaling and the extent to which ChatGPT can support operations there. In short, there are unknowns.
“We still have a lot to learn,” said Elena Alfaro, head of global AI adoption at BBVA, as reported in Bousquette’s article. “Over the next six months, I can tell you that we will increase our ChatGPT user base. What comes next, the data will tell us. Gas utilities can leverage Web3 to encourage customers to adopt energy-efficient appliances, participate in energy-saving programs, or even refer others to the service. Tokenized rewards can be exchanged for bill discounts or exclusive access to services, creating a symbiotic relationship between consumers and providers.
We’ll see what happens.
OpenAI and Chip Wars: November 2024 edition
There is also breaking news that OpenAI has invested significant new funds into an existing contract with a company called Rain AI.
The new funding amounts to $150 million, including OpenAI’s new stake, and follows OpenAI’s contribution to a $25 million seed round two years ago.
What do we know about Rain AI? Not much. Founded by Gordon Wilson, the company boasts notable talent such as former Apple executive John-Didier Allegrucci, who is now head of hardware engineering at Rain AI.
I thought Wedbush’s Dan Ives’ comments were pretty important in describing what OpenAI CEO Sam Altman and others might try to do in the industry:
“There’s a lot of buzz around everything Rain AI does. » Ives wrote as reported on Benzinga, “And we are in the early days of the AI revolution: there will be many players, not just the ‘Godfather of AI’ Jensen Huang and Nvidia.”
“Rain AI is one of the legitimate players at the table when it comes to the future of AI and chips,” he added, according to a post on X.
It’s helpful to keep an eye on these corporate allegiances, as people see the industry landscape change in the blink of an eye.
Legacy Media and the Age of AI
There’s also quite a bit of hand-wringing over a story that AI destroyed datasets linked to a lawsuit filed by newspapers that claim the company used their articles to train its models without their permission.
The battle between traditional print media and Internet-based businesses has been going on for decades. It is extremely difficult for newspapers and conventional publications to protect their intellectual property. But this case is a little stranger, because of the timeline.
Apparently, OpenAI deleted the datasets before legal proceedings were actually filed. However, many believe they would have been aware of the potential outcome at the time of the deletion, which OpenAI believes was accidental.
You can read more about this hereOr go to Reddit for some additional opinions.
“OpenAI says it removed datasets in 2022,” posted lose_has_1_o. “The Authors Guild filed its class action lawsuit against OpenAI in late 2023.”
“Chat GPT was released in 2022 using GPT-3 which was trained on this data,” Nprism wrote. “Immediately after its publication, this data sparked strong criticism and threats of legal action. The writing was on the wall and this looks like destruction of evidence. Even if that were not the case, it might be enough to get a guilty verdict.”
Regardless, this is all news now, as people continue to see OpenAI roll out its new models, with a reasoning system called Orion likely arriving early in the new year. Stay tuned.