Sponsored by |
 |
|
|
Meta just launched Muse Image, and it's already the second-best image generator on the internet by public vote - trailing only OpenAI's GPT Image 2. The bigger story might be where it's headed next: straight into Instagram, WhatsApp, and soon Facebook, putting a serious AI art tool in front of nearly every Meta user on the planet. |
Benchmark scores aside, is raw image quality even the point anymore, or has the real AI art race shifted to whoever owns the biggest distribution pipe? |
Today in AI: |
Meta's Muse Image lands on Instagram and WhatsApp xAI's Grok 4.5 trains directly inside Cursor OpenAI's GPT-5.6 Sol goes generally available |
How Jennifer Aniston's LolaVie brand grew sales 40% with CTV ads |
|
The DTC beauty category is crowded. To break through, Jennifer Aniston's brand LolaVie, worked with Roku Ads Manager to easily set up, test, and optimize CTV ad creatives. The campaign helped drive a big lift in sales and customer growth, helping LolaVie break through in the crowded beauty category. |
Learn more |
|
|
|
What's new? Meta launched Muse Image, a new AI image generator from Meta Superintelligence Labs that landed the #2 spot on Arena's text-to-image leaderboard, trailing only OpenAI's GPT Image 2 - and it's rolling out through Meta AI, Instagram Stories, and WhatsApp, with Facebook next. |
What matters? |
Muse Image beat both Google and Microsoft's top image models on the leaderboard, and its companion Muse Video model grabbed #3 in Arena's text-to-video rankings. Users can @-mention Instagram accounts to pull public photos directly into a generation, and the tool can incorporate trending content for inspiration. The model is already wired into Advantage+, Meta's ad-automation suite that drives roughly $60 billion in annualized revenue, letting advertisers generate creative straight from a campaign brief. |
Why it matters? |
Image quality is no longer the clearest differentiator - distribution is. By building Muse Image directly into Instagram, WhatsApp, and its ad tools, Meta is betting reach beats raw benchmark scores. |
|
GUIDE |
|
|
|
What's new? xAI released Grok 4.5, its strongest model yet, built specifically for coding and agentic work and trained directly alongside Cursor using real developer session data. |
What matters? |
The model was benchmarked using real Cursor sessions rather than synthetic tests, measuring what it can actually get done inside a live codebase over a long stretch of work. Grok 4.5 is available today in Grok Build, inside Cursor itself, and through the SpaceXAI console, putting it directly in front of working developers rather than just chatbot users. The much-anticipated Grok 5, xAI's rumored 6-trillion-parameter model, is still training - Grok 4.5 is the interim flagship while that larger release slips past its original window. |
Why it matters? |
Coding has become the proving ground for agentic AI, and xAI is betting that real developer session data beats a bigger parameter count. Whether that holds up once Grok 5 finally ships is the open question. |
|
SPONSORED BY ELEVEN LABS |
The best voice models, now with full orchestration. Build real-time voice and chat agents on one low-latency stack: any LLM, your tools and knowledge, testing, Guardrails, and omnichannel deployment. |
Start building free |
|
|
|
|
What's new? OpenAI rolled out GPT-5.6 Sol to general availability across ChatGPT, Codex, and the API, its flagship model in a new three-tier family that also includes Terra and Luna. |
What matters? |
Sol tops the family with an 88.8% score on Terminal-Bench 2.1, while a specialized "Sol Ultra" subagent mode pushes that to 91.9%, with Terra and Luna offering cheaper, faster alternatives. Context window climbs to roughly 1.5 million tokens, a jump aimed at longer, multi-hour agentic tasks like coding and research. Pricing lands at $5 per million input tokens and $30 per million output tokens, undercutting rivals while claiming comparable frontier-level performance. |
|
Why it matters? |
A cheaper, longer-context flagship model raises the bar for every team building agentic tools on top of frontier AI. With Sol now generally available, the "model wars" become less about preview access and more about which one developers actually pick. |
|
Everything else in AI |
Runway introduced Agent 2.0, an upgraded marketing agent that turns a single prompt into a full campaign - briefs, visual and video assets, and performance analysis included. |
Google patched a "Rogue Agent" flaw in Dialogflow CX that could have let attackers hijack AI chatbot conversations and exfiltrate data before researchers caught it. |
Higgsfield added Gemini Omni Flash and Seed Audio 1.0 to its Adobe Premiere Pro and DaVinci Resolve plugins, pushing further into professional editing workflows. |
|
Essential AI Guides - Reading List: |
|
|
Let us know! |
What did you think of today's email?
Before you go, please give your feedback to help us improve the content for you!
|
|
|
Work with us |
Reach 100k+ engaged Tech Professionals, Engineers, Managers and decision makers. Join brands like MorningBrew, HubSpot, Prezi, Nike, Ahref, Roku, 1440, Superhuman, and others in showcasing your product to our audience. Get in touch now → |
|
