One of the biggest benefits of Brad, an experiment to collaborate with generative AI, is that it can tailor its responses to exactly what you need. For instance, you could ask Bard to start a trip planning Doc for you and your friends, draft up your online marketplace listing, or help explain a science topic to your kids. And now, Bard is getting even better at customising its responses so you can easily bring your ideas to life.
Today we’re rolling out Bard’s most capable model yet. Bard now integrates with Google apps and services for more helpful responses. We’ve also improved the “Google it” feature to double-check Bard’s answers and expanded features to more places.
Double-check information you get from Bard
Bard has tools to help you identify potentially inaccurate statements. One way to double-check Bard’s responses is to use the Google button. This uses Google Search to find content that helps you assess and further research the information you get from Bard.
Bard’s double-check feature can make mistakes
For example, the feature may show that Google Search found content that makes a similar statement to Bard’s. But the content may actually contradict Bard. The web content may be inaccurate, too. You should read, review, and carefully evaluate the content identified by the double-check feature, as well as its context.
Help make Bard better
Google relies on people to give feedback on answers that don’t seem right. That’s why Bard has built-in features to collect your feedback.
- Mark good responses
- Mark bad ones
- Use the Report legal issue button where appropriate
Be mindful of how you use Bard
You have an opportunity to shape Bard for others. Bard will use the conversations you have to improve future responses for everyone so don’t enter anything you wouldn’t want a reviewer to see or Google to use.
How and when does Bard cite sources in its responses?
Bard, like some other standalone LLM experiences, is intended to generate original content and not replicate existing content at length. We’ve designed our systems to limit the chances of this occurring, and we will continue to improve how these systems function. If Bard does directly quote at length from a webpage, it cites that page. For answers with URLs or image thumbnails, Bard enables users to easily see and, in some cases, click to navigate directly to the source for each. If Bard generates a response using content from an extension, like Google Workspace, it provides a link to the sources (like your emails) that it used.
Sometimes the same public content may be found on multiple webpages and Bard attempts to point to a popular source. In the case of citations to code repositories, the citation may also reference an applicable open source license.
Bard is an experiment, and we’ll continue to learn, iterate, and improve the experience as we get feedback from a range of stakeholders including people like you, publishers, creators, and more.
Is Bard able to explain how it works?
LLM experiences (Bard included) can hallucinate and present inaccurate information as factual. One example is that Bard often misrepresents how it works. We’ve seen this occur in a number of instances—for example, in response to prompts asking how it was trained or how it carries out various functions (like citing sources, or providing fresh information).
Bard responses may also occasionally claim that it uses personal information from Gmail or other private apps and services to train Bard’s generative machine-learning technologies. That’s not accurate, and as an LLM interface, Bard does not have the ability to determine these facts. We do not use personal data from your Gmail or other private apps and services to train Bard’s generative machine-learning technologies. More information about how we keep your data private, safe.
These are some of the reasons we released Bard as an experiment and want to be as transparent as possible about the limitations of LLMs, including providing disclaimers within the Bard experience. If you see a hallucination or something that isn’t accurate, please click the thumbs down button and provide your feedback. That’s one of the ways Bard will learn and improve.
Can Bard help with coding?
Yes, Bard can help with coding and topics about coding, but Bard is still experimental and you are responsible for your use of code or coding explanations. So you should use discretion and carefully test and review all code for errors, bugs, and vulnerabilities before relying on it. Code may also be subject to an open source license and Bard provides related information. Read more about how and when Bard cites sources .
Who can use Bard?
To use Bard, you must be 18 or over. You also need a personal Google Account that you manage on your own, or a Google Workspace account for which your admin enabled access to Bard. If you’re a Google Workspace admin.
You can’t access Bard with a Google Account managed by Family Link or with a Google Workspace for Education account designated as under the age of 18.
Right now, Bard may not be available in your country, but we will be rolling out to new languages and countries over time.
How many languages does Bard support?
Bard is available in more than 40 languages, including: English, Japanese, Korean, Arabic, Bahasa Indonesia, Bengali, Bulgarian, Chinese (Simplified / Traditional), Croatian, Czech, Danish, Dutch, Estonian, Farsi, Finnish, French, German, Gujarati, Greek, Hebrew, Hindi, Hungarian, Italian, Kannada, Latvian, Lithuanian, Malayalam, Marathi, Norwegian, Polish, Portuguese, Romanian, Russian, Serbian, Slovak, Slovenian, Spanish, Swahili, Swedish, Tamil, Telugu, Thai, Turkish, Ukrainian, Urdu and Vietnamese. We are continuing to teach Bard how to respond in even more languages.
DN Review