Bard , Google ‘s chatbot that aims to compete with ChatGPT and Bing , has been criticized by several users, but Sundar Pichai says updates are on the way.
Sundar Pichai , the current chief executive of Google , responded to criticism from artificial intelligence enthusiasts against Bard , the firm’s experimental chatbot, promising that updates would be released soon to improve its performance. The CEO indicated that new versions of the platform – which aims to compete against OpenAI ‘s ChatGPT and Microsoft ‘s Bing – are already being developed and are on the way.
Through an interview with the Hard Fork podcast of The New York Times , the businessman revealed new details regarding the important Google project . “Clearly we have more capable models. Very soon, perhaps while this podcast airs, we’ll be updating Bard with some of our more capable PaLM models , bringing more capabilities, whether it’s in reasoning, coding, or better answering math questions. So you’ll see the progress over the next week , ” Pichai said .
Bard, Google’s chatbot, will improve soon
Pichai mentioned that Bard is powered by a “lightweight and efficient version of LaMDA ,” an AI language model that focuses on delivering dialogue. “In a way, I feel like we’ve taken an upgraded Civic and put it up against more powerful cars ,” the CEO said. On the other hand, PaLM – a more recent language model that is also larger in scale – is more capable when it comes to tasks like common-sense reasoning and coding problems, Google claims .
Bard was made available to the public on March 21, but it did not attract the attention or good reviews of ChatGPT and the Bing chatbot as, during their first tests, experts found that this platform was consistently less useful than its competitors. Like all general-purpose chatbots, Bard is capable of answering a wide range of questions, but its answers tend to be less fluid and imaginative, and not based on reliable data sources.
Pichai suggested that part of the reason for Bard’s limited capabilities was a sense of caution within Google . “It was important to me that we don’t release a more capable model before we’re absolutely sure we can handle it well ,” she said. In addition, the businessman also confirmed that he was speaking with Larry Page and Sergey Brin , co-founders of Google, about the project, stating that he never issued the infamous “code red” to boost the development of AI in the face of the popularity of ChatGPT despite the fact that “There were probably people in the company who sent emails saying there’s a code red . “
Google is still concerned about the rapid advance of AI
Regarding the debate on whether AI poses a threat to today’s society due to its rapid progress, Pichai said the following: “In this area, I think it is important to listen to concerns and I think there are merits to worry about it. This is going to need a lot of debate, no one knows all the answers, no company can get it right. AI is too important an area not to regulate .” However, the businessman also noted that it was better to just apply regulations in already existing sectors than to create new laws to specifically address AI .
The truth is that there are still a number of concerns about chatbots , such as their tendency to spread false information or even those that suggest that these systems are so difficult to control that, once connected to the wider network, they could be used in harmful ways. Others indicate that current programs are also approaching what is known as artificial general intelligence : systems that are at least as capable as a human at a wide range of tasks.
“It’s very clear to me that these systems are going to be very, very capable, so it almost doesn’t matter whether it’s gotten to artificial general intelligence or not. Can we have an AI system that can cause disinformation at scale? Yes. Is it artificial general intelligence? It doesn’t really matter. Why do we need to worry about AI security? Because you have to anticipate it and evolve to meet that moment,” Pichai said during the conversation.