As Google’s interactive AI tool Bard approaches its first anniversary, it continues to evolve. Initial feedback highlighted Bard’s lack of human-like responses, but Google has addressed this by integrating its AI Gemini model, allowing Bard to extend its capabilities to video, audio, and images. In a recent blog post, Jack Krawczyk, Bard’s product director, announced that the tool now offers free image creation. Users can simply request an image, and Bard will generate customized multi-images to bring their ideas to life. However, the image generation process takes approximately 13 seconds. I tested Bard by asking it to create an image for a news article, but it refused. When I asked it to create a simple image of a blonde cyborg, that’s exactly what it produced. Over the past year, numerous chatbots have emerged, including Google’s Bard, OpenAI’s ChatGPT, Microsoft’s Bing AI, and Anthropic’s Claude, as major tech companies compete for dominance in this rapidly-growing field. These chatbots leverage large datasets and diverse languages to offer voice, and now image and video, responses to user queries. While they can provide confident answers, they are not always accurate and are susceptible to misuse. Bard also differs from the original human artwork by adding watermarks to created images. I tested this by asking Bard to replicate Botticelli’s “The Birth of Venus,” and while the result was similar, it was noticeably different in quality. Bard also rejects requests to create violent, offensive, or pornographic content, and uses filters to prevent the generation of images featuring famous individuals. When asked to create an image of a specific journalist, Bard claimed it did not have enough information. Despite this, it successfully produced an image of the journalist at work with two sandwiches on the table. However, Bard refused to create an image of someone throwing money from the Hoover Dam due to legal reasons. On the other hand, it readily generated images of historical events such as the signing of the Declaration of Independence. Google has also expanded Bard’s Gemini Pro availability from English to over 40 languages, intending to reach more than 230 countries and territories. This expansion was prompted by the addition of Gemini Pro to Bard in December 2023, enhancing its understanding, reasoning, summarization, and writing skills. It’s worth noting that CNET is utilizing an AI engine to assist in generating news content. For further details, refer to this post.
