OpenAI has recently unveiled an upgrade for ChatGPT (available on Apple and Android platforms) featuring two noteworthy additions: AI voice options, allowing users to hear the chatbot’s responses, and image analysis capabilities. The image function bears resemblance to the existing offering from Google’s Bard chatbot.
After extensive testing to push the boundaries of ChatGPT’s capabilities, OpenAI’s chatbot continues to both astonish and concern me simultaneously. While I was genuinely impressed with the web browsing beta introduced through ChatGPT Plus, my apprehensions persisted, particularly regarding its implications for online content creators and various other concerns. The introduction of the new image feature for OpenAI subscribers left me with a similarly ambivalent sentiment.
Although I haven’t had the opportunity to experiment with the new audio capabilities (other accomplished reporters on our team have), I did get a chance to assess the impending image features. Here’s a guide on accessing and utilizing the upcoming image search functionality in ChatGPT.
How to Utilize ChatGPT’s Image Features
While the update is anticipated to launch before the year’s end, the exact release date for the image and voice features remains uncertain. As is customary with most of OpenAI’s updates, such as the GPT-4 version of ChatGPT, paying subscribers are granted priority access.
There are three methods for uploading photos within the ChatGPT mobile app. First, you can select the camera icon located to the left of the message bar and capture a new photo using your smartphone. Before uploading the image, you can use your finger to draw a circle around the specific subject you want the chatbot to focus on. Alternatively, you can choose photos from your device’s gallery for upload, including files saved on your phone. For desktop browser users of ChatGPT, uploading saved photos from their computer is an option. While there is presently no capability to upload videos to the chatbot, it is possible to submit multiple images within a single prompt.
It’s worth noting that ChatGPT does have its limitations. When presented with a random photo of a mural, it was unable to identify the artist or location. However, it readily recognized the locations of various San Francisco landmarks in images, such as Dolores Park and the Salesforce Tower. While it may still appear somewhat gimmicky, individuals exploring new cities, countries, or neighborhoods might find some enjoyment in experimenting with ChatGPT’s visual features.
