Top stories in AI today: – New Google Pixel lineup goes big on AI – NASA, IBM launch AI to decode the sun
– Use GPT-5 in Microsoft 365 to analyze emails
– Gemini expands to the home with Nest
– 4 new AI tools, community workflows, and more Read more: https://
therundown.ai/p/googles-pixe
l-10-lineup-goes-big-on-ai
…
MULTIMODAL AI
-
Google, NASA, and Microsoft expand AI across devices and services
By
–
-
New AI Model Generates Emotive Japanese Singing in Seconds
By
–
A full minute of emotive, pristine singing—in Japanese, no less. And it's able to be generated in less than 6 seconds with our new model.
— Pika (@pika_labs) 20 août 2025
We're pretty impressed with @XVisualneuFX (and ourselves) pic.twitter.com/gBW6i8EbtMA full minute of emotive, pristine singing—in Japanese, no less. And it's able to be generated in less than 6 seconds with our new model. We're pretty impressed with @XVisualneuFX (and ourselves)
-
Google Photos Introduces AI-Powered Image Editing with Voice Commands
By
–
Edit images in Google Photos by simply asking
— Chubby♨️ (@kimmonismus) 20 août 2025
Pretty cool feature. Your turn, @Apple pic.twitter.com/Z60Sjf1tRWEdit images in Google Photos by simply asking Pretty cool feature. Your turn, @Apple
-
Character.ai Announces Major Summer Updates and New Features
By
–
SO.MANY.UPDATES. It’s been a packed summer at http://
c.ai and we wanted to give you the full story of what the team has been up to! : -
ElevenLabs v3 Launches on Poe with Advanced Audio Features
By
–
ElevenLabs v3 is now on Poe!
— Poe (@poe_platform) 20 août 2025
With new audio tags, multi-speaker support, and options for over 70 languages, this category leading model gives you more control to create audio that feels more real, natural and expressive. (1/2) pic.twitter.com/oZ8ZOL4chqElevenLabs v3 is now on Poe! With new audio tags, multi-speaker support, and options for over 70 languages, this category leading model gives you more control to create audio that feels more real, natural and expressive. (1/2)
-
ComfyUI Workflows Transform from Complex to One-Shot Results
By
–
Los complejos workflows de ComfyUI de hace un año se acaban convirtiendo en resultados inmediatos (one-shot) un año después. La IA imparable.
-
Qwen Edit: AI-Powered Text-to-Image Editing Model Released
By
–
introducing Qwen Edit.
— KREA AI (@krea_ai) 20 août 2025
this new models offers incredible image editing capabilities from text prompts.
try it now for free! pic.twitter.com/sQzgs503KIintroducing Qwen Edit. this new models offers incredible image editing capabilities from text prompts. try it now for free!
-
V-JEPA 2 Learns World Models from Million Hours Video
By
–
internet videos to robot actions? @AIatMeta
's V-JEPA 2 learns a world model from over 1 million hours of video, enabling it to perform complex, real-world robotic tasks zero-shot. Join author Nicolas Ballas for a talk on V-JEPA 2 this Friday! -
Effective Training Data Synthesis for Improving MLLM Chart Understanding
By
–
Effective Training Data Synthesis for Improving MLLM Chart Understanding
-
Creating Multimodal AI Content: Song, Image, Video Generation Experiment
By
–
This was obviously for fun, and not a serious proposal, but I find doing little experiments is a good way to understand the state of AI and its limits It took about 15 minutes. I generated the song with Suno, the initial image with Midjourney, Veo 3 for videos, Sync for lip sync