Major Google Bard update allows it to not just write code, but execute it
Latest version of ChatGPT rival sees maths and coding accuracy improve by nearly a third
Your support helps us to tell the story
From reproductive rights to climate change to Big Tech, The Independent is on the ground when the story is developing. Whether it's investigating the financials of Elon Musk's pro-Trump PAC or producing our latest documentary, 'The A Word', which shines a light on the American women fighting for reproductive rights, we know how important it is to parse out the facts from the messaging.
At such a critical moment in US history, we need reporters on the ground. Your donation allows us to keep sending journalists to speak to both sides of the story.
The Independent is trusted by Americans across the entire political spectrum. And unlike many other quality news outlets, we choose not to lock Americans out of our reporting and analysis with paywalls. We believe quality journalism should be available to everyone, paid for by those who can afford it.
Your support makes all the difference.Google has unveiled a major new update to its AI chatbot Bard that significantly improves its powers of logic and reasoning.
The latest version of the ChatGPT rival is now capable of both writing and executing code by itself, the tech giant announced, allowing it to figure out problems on a far deeper level than current generative AI systems.
Google’s artificial intelligence tool is able to perform the new tasks through a new technique called “implicit code execution”, which enables Bard to detect computational prompts and run code in the background.
The result is that Bard should theoretically be able to respond more accurately to mathematical tasks and coding questions, as it will have already tested the outcomes that it proposes.
Until now, large language models (LLMs) like ChatGPT and Bard have been better suited to language and creative tasks, as they draw from their training data to predict what word will come next when talking about a specific subject.
This allows them to produce text quickly but without deep thought, making them weaker when it comes to areas like reasoning and mathematics.
“Our new method allows Bard to generate and execute code to boost its reasoning and maths abilities,” Google wrote in a blog post on Wednesday.
“With the latest update, we’ve combined the capabilities of both LLMs and traditional code to improve accuracy in Bard’s responses. Through implicit code execution, Bard identifies prompts that might benefit from logical code, writes it ‘under the hood,’ executes it and uses the result to generate a more accurate response.”
The new method improved Bard’s accuracy for coding and maths problems by roughly 30 per cent during internal tests, Google claimed.
Accuracy remains one of the biggest issues with AI chatbots, with Google warning that despite the upgrade, Bard “won’t always get it right”.
Unreliable or fabricated information generated by these AI tools is known as hallucinations, and they are typically delivered in a confident way that can be even more misleading for the user.
ChatGPT creator OpenAI announced a potential new method to improve AI misinformation last month, involving two AI systems debating each other until they agree on the correct answer.
Join our commenting forum
Join thought-provoking conversations, follow other Independent readers and see their replies
Comments