ChatGPT-rival Bard AI Accidentally Reveals It Relays On Gmail Data

Google’s Artificial Intelligence (AI) assistant Bard, designed to compete with ChatGPT, has caused controversy due to its alleged claims that it was trained on users’ Gmail data.

Bard was created to give competition to ChatGPT, which is based on the GPT-3.5, for free users and GPT 4 architecture, and has similar functionalities. However, privacy seems to be of concern around Google’s AI chatbot and the data set it uses.

The controversy surrounding Bard was brought into the limelight by Microsoft researcher Kate Crawford, who shared a screenshot of a conversation with the chatbot.

Crawford underlined that Bard listed publicly available datasets from sources like ‘Wikipedia and GitHub’, and other internal data from Google products including Gmail and third-party companies when asked about its dataset.

She tweeted, "Anyone a little concerned that Bard is saying its training dataset includes... Gmail? I'm assuming that's flat out wrong, otherwise Google is crossing some serious legal boundaries.”

Meanwhile, Google has completely denied the Crawford’s claims, stating Bard to be an “early experiment” and that is not trained on Gmail data.

Crawford also shared that Google had responded to her tweet mentioning there was no usage of private data in training. https://twitter.com/katecrawford/status/1638325816844009474

However, the incident has highlighted the limited features of both generative AI tools, Bard and ChatGPT, which have a long way to go.