[ad_1]
The information: Google DeepMind has launched a brand new watermarking device which labels whether or not footage have been generated with AI. The device, known as SynthID, will permit customers to generate photos utilizing Google’s AI picture generator Imagen, then select whether or not so as to add a watermark.
Why watermarking? In the previous yr, the massive recognition of generative AI fashions has introduced with it the proliferation of AI-generated deepfakes, non-consensual porn, and copyright infringements. Watermarking—a method the place you disguise a sign in a bit of textual content or a picture to establish it as AI-generated—has change into some of the common coverage options to curb harms.
Why it issues: The hope is that SynthID might assist folks establish when AI-generated content material is being handed off as actual to counter misinformation, or assist shield copyright. Read the total story.
—Melissa Heikkilä
Interested within the impression of generative AI? Read extra about this matter:
+ These new instruments might assist shield our footage from AI. PhotoGuard and Glaze are simply two new programs designed to make it more durable to tinker with pictures utilizing AI instruments. Read the total story.
+ AI fashions spit out pictures of actual folks and copyrighted photos. The discovering might strengthen artists’ claims that AI corporations are infringing their rights. Read the total story.
+ These new instruments allow you to see for your self how biased AI picture fashions are. DALL-E 2 and two latest variations of Stable Diffusion have a tendency to provide photos of those who look white and male, particularly if the immediate is a phrase like ‘CEO’. Read the total story.
The must-reads