Google’s Ai Transforms Text Into Images With Remarkable Realism
Google Research has developed an advanced artificial intelligence (AI) system that can turn any phrase into a strikingly realistic photo.
FREMONT, CA: Google’s AI is capable of creating photorealistic images with an unprecedented degree of photorealism and a profound level of language understanding. Translating short text descriptions into visuals is not a groundbreaking idea. Earlier this year, OpenAI's Dall-E2 AI exhibited the capacity to create graphics from a brief description and then fine-tune the outcome using a minimal set of tools. While the "photorealism" of those photos was outstanding, the system was most impressive for its ability to create outcomes in a variety of artistic styles. Google's method appears to be more focused on making images look like real photos, and the samples shared were made with the system look considerably more like photos than OpenAI's drawings.
Direct access to the AI is not available to the public, as it was with OpenAI's system, because Google does not believe it is ready yet; all of the examples are pre-generated. Google's system, like others, is trained on enormous volumes of data gathered from the internet that are not curated, which might cause issues and could be misused if released to the public in its current state. The researchers noted that while this method facilitated significant algorithmic breakthroughs in recent years, these datasets frequently reflect social preconceptions, authoritarian perspectives, and disparaging or otherwise detrimental linkages to minority identity groups.
The LAION-400M dataset, which is known to contain a wide range of inappropriate content, including pornographic imagery, racist slurs, and harmful social stereotypes, while a subset of our training data was filtered to remove noise and undesirable content, such as pornographic imagery and toxic language was employed. The researchers discovered that AI already has social prejudices and tends to construct representations of people with lighter skin tones and place them in stereotyped gender roles. The researchers would investigate a framework for responsible externalization in future work that balances the use of external auditing with the concerns of unrestrained open access.