- AI Valley
- Posts
- A mysterious new image generation model
A mysterious new image generation model
PLUS: Meta set to debut new smart glasses for under $1,000
Together with
Howdy. It’s Barsee again.
Happy Monday, AI family, and welcome back to AI Valley.
Today’s climb through the Valley reveals:
Meta set to debut new smart glasses for under $1,000
Sam Altman explores trillion-dollar roadmap after GPT-5
A mysterious new image generation model from Google
Anthropic gives Claude the ability to exit abusive conversations
Plus trending AI tools, posts, and resources
Let’s dive into the Valley of AI…
MOVIEFLO AI
MovieFlo.AI is where AI video meets professional storytelling: one platform for studios, marketers, and creators who think in scenes, not prompts.
In one smooth workflow, you can go from idea to final video with built-in tools that help you cast actors, plan scenes, and generate cinematic results.
Top video models like Veo, Kling, Wan, and Runway are right at your fingertips: no setup, no guesswork.
Built by veterans from LucasFilm and ILM, MovieFlo.AI delivers professional-grade tools with consistency baked in: from actors and props to branding and animation.
From product ads to short films, this is your production studio in a browser.
🎬 Try it for free, with 10% off annual subscriptions, exclusively for AI Valley readers → https://bit.ly/MovieFloAI
*This is sponsored
THROUGH THE VALLEY
Meta is preparing to launch its first smart glasses with a built-in display, codenamed Hypernova, in September at around $800 (far less than the $1,000–$1,400 price many expected). Instead of a full AR headset, the glasses serve as a middle step, offering a small display in the right lens for quick access to notifications, apps, and alerts. They’ll also work with Meta’s neural wristband, which lets users control the glasses with subtle hand movements. Positioned against Apple’s $3,499 Vision Pro, Meta is betting on affordability and practicality to bring AR closer to everyday use.
OpenAI employees may soon get a big payout. A new deal being discussed would let current and former staff who spent at least two years at the company sell about $6 billion worth of shares to investors, including Thrive Capital, SoftBank, and Dragoneer. If it goes through, the deal would value OpenAI at $500 billion. Talks are still in the early stage, so details could change, but the move would give employees a chance to cash out as competition for top AI talent continues to rise.
OpenAI recently hosted a dinner with reporters in San Francisco, where Sam Altman and his team talked less about GPT-5 and more about the company’s bigger ambitions. GPT-5’s launch drew mixed reviews, with many users complaining about its colder tone and missing older models like GPT-4o. Altman admitted they mishandled the rollout and promised clearer transitions in the future, while updates are already being made to make GPT-5 “warmer.” Despite the criticism, OpenAI says API traffic doubled within 48 hours of launch, showing strong demand. Beyond models, Altman confirmed plans for new consumer apps under incoming executive Fidji Simo, hinted at an AI-powered browser, a possible social platform, and even investments in brain-computer interfaces. With big bets in hardware, data centers, and robotics, it’s clear OpenAI is positioning itself to grow far beyond ChatGPT and possibly head toward an Alphabet-style future.
A mysterious new image generation model called “Nano Banana” has appeared on Reddit’s AI testing platform LMArena, sparking speculation that it may be a secret Google project. The model produces highly realistic and detailed images, can edit existing photos with simple prompts, and even rebuilds entire scenes instantly—without relying on layers like traditional editing tools. It also preserves character consistency across comics or ads, while supporting artistic transformations from watercolor to photorealism. Early tests show it outperforms existing models in tasks like sharpening blurry images or merging photos, though some details appear illogical. While unconfirmed, if “Nano Banana” is indeed from Google, it could disrupt the image-generation race once again.
Anthropic has added a new safety feature to its Claude Opus 4 and 4.1 models that allows them to end chats if conversations become abusive or harmful. The system activates after attempts at redirection fail, especially around topics like minors, terrorism, or violence. Tests showed the models could even “stop” abusive exchanges on their own, though users can instantly start a new chat or edit past messages. Importantly, the cutoff won’t trigger during self-harm or crisis situations, making this one of the first steps toward AI “wellness.”
TRENDING TOOLS
Agent Opus - The first AI video agent built for social media
Higgsfield Product-to-Video - Drop in product images and instantly generate fully animated, cinematic shots
Snowglobe - Simulate real users to test your AI before you launch publicly
Perplexity Comet - A next-gen AI browser that helps you navigate the web faster
Continua - Joins your group chats to track decisions, answer with live web search, and turn conversations into shared docs
Notion - Offline mode is finally available
AI Bank Statement Analyzer - Transform PDF bank statements into queryable financial insights
Prophet Arena - The AI benchmark for general predictive intelligence
THINK PIECES / BRAIN BOOST
GPT-5 coding cheat sheet
35 thoughts about AGI and 1 about GPT-5
One prompt to master prompt engineering
ChatGPT’s mobile app has generated $2B to date and earns $2.91 per install
Building a web search engine from scratch in two months with 3 billion neural embeddings
The “Palantir Mafia” behind Silicon Valley’s hottest startups
This GitHub is an AI agent goldmine: RAG-powered assistants, MCP-integrated agents, deep research agents
THE VALLEY GEMS
What’s trending on social today:
1/ AI progress since GPT-3.5 release.
AI Progress since GPT-3.5
OpenAI seems to be slowing down with GPT-5
Anthropic incredibly steady progress
Google had it's breakthrough with Gemini 2.5 Pro— Lisan al Gaib (@scaling01)
7:50 PM • Aug 17, 2025
2/ Half of these have been obsolete careers for decades.
Microsoft, $MSFT, has said that these are the 40 jobs most at risk by AI
— unusual_whales (@unusual_whales)
6:01 PM • Aug 17, 2025
3/ Andrew Garfield has been cast as Sam Altman for the OpenAI movie ‘Artificial’.
Omg Andrew Garfield is reading my book for the OpenAI movie. reddit.com/r/OpenAI/comme…
— Karen Hao (@_KarenHao)
1:12 PM • Aug 16, 2025
4/ China just hosted the first-ever World Humanoid Robot Games, bringing together robots from 16 countries to compete in sports and real-world tasks like customer service.
THAT’S ALL FOR TODAY
Thank you for reading today’s edition. That’s all for today’s issue.

💡 Help me get better and suggest new ideas at [email protected] or @heyBarsee
👍️ New reader? Subscribe here
Thanks for being here.
HOW WAS TODAY'S NEWSLETTER |
REACH 100K+ READERS
Acquire new customers and drive revenue by partnering with us
Sponsor AI Valley and reach over 100,000+ entrepreneurs, founders, software engineers, investors, etc.
If you’re interested in sponsoring us, email [email protected] with the subject “AI Valley Ads”.