The World Economic Forum at Davos has become the latest stage for discussions on artificial intelligence, where the topic has been labeled a key issue in this year’s gathering.
Sam Altman, the CEO of OpenAI, is become one of the most prominent names in attendance at Davos, along with other high-profile tech CEOs.
In a Wednesday interview, Altman said GPT5, the upcoming version of his firm’s large language model, will invariably be able to do a lot more than previous iterations, including the groundbreaking GPT4, which is the engine behind ChatGPT.
“The thing that matters most is not that it can have this new modality or it can solve this new problem, it is that the generalized intelligence keeps increasing,” said Altman in an interview with Axios.
“That is the high-ordered bit,” he said. “That is, the overall capabilities of the model, its overall intelligence. Its ability to do longer, more complex problems, more accurately, more of them.”
Altman says he now understands that the “productization” of AI is now a critical step in the process of his company, which came to be as a research company but now increasingly has relied on product development as AI increasingly becomes adopted by users in their daily lives.
AI Assistants Are Here To Stay In 2024
Natural language AI models, that users can talk to or chat via text using everyday language, will likely increase their speed this year. This will allow for a surge in human-computer interaction via natural language, Altman said.
What this means is that more and more people will begin to use their computers with their voice to perform complex tasks instead of having to tap on a screen or use a mouse.
Last week, a new device was unveiled at CES 2024 that is aiming to replace the smartphone experience with an AI assistant that can perform most daily tasks normally relegated to visual interfaces, like calling a car ride, answering a text message or looking for a food recipe.
“We’re heading towards this new way to do knowledge work,” says Altman. “I think we’ll see people do more and more of their workflow inside of a language model.”
Microsoft Corp MSFT is already working toward this strategy. The company recently saw its search engine market share rise by 3.4% after its $13-billion partnership with OpenAI started bearing fruit.
Now, Microsoft announced it’s releasing a subscription-based access to AI Copilot, its AI assistant service, for $20 a month for individual consumers and small businesses.
Investors are keenly looking at Microsoft ahead of its second-quarter earnings call at the end of the month.
On Wednesday, Alphabet Inc GOOG GOOGL announced the latest generation of Samsung devices carrying Android will include an onboard version of the company’s AI engine, Gemini. The product will allow users to perform daily tasks using Gemini as an AI assistant.
Copyright Infringement And Impact Of AI In Media
Altman emphasized his position over an ongoing lawsuit with the New York Times Co NYT. Last month, the media company sued OpenAI and Microsoft for “billions of dollars in statutory and actual damages” over alleged copyright infringements in the use of its content to train AI models.
“We’re happy to include NY Times training data if the NY Times would like and there and many other partners who want us to, but we don’t need to. In fact as these models get smarter and better at reasoning, we need less training data,” said Altman.
Publishers like The Atlantic are also raising concerns that the integration of media content into AI chatbots will rob traffic from their sites, making their business model less sustainable.