5 min read

🛎️ GPT Image 1.5 Is Coming

Plus: Zoom Launched AI Companion 3.0, AI Rewrites Cashmere

Good Morning, AI Enthusiasts!

AI is moving out of demos and into the control layers of creation, coordination, and industrial supply.



NEW LAUNCH

OpenAI Updates Images, Gap Narrows Slightly

OpenAI has released GPT Image 1.5 inside ChatGPT and via API, paired with a redesigned image workspace and preset driven workflows. The upgrade focuses on faster generation, easier style transforms, and more controllable image edits rather than a headline grabbing model jump. Even the 1.5 label signals caution. This is not a reset or a leapfrog moment, but a measured iteration aimed at closing distance with Google’s Banana Pro after months of relative silence.

In practice, GPT Image 1.5 improves day to day usability more than peak quality. Presets and conversational image editing lower friction for non expert users and speed up routine tasks like portraits, product shots, and simple scene edits. However, in areas that matter for production reliability such as text accuracy, layout discipline, and typography, Banana Pro still appears more consistent. Teams choosing today will likely optimize for workflow versus precision, not novelty.

GPT Image 1.5 is a solid catch up release, but Banana Pro still sets the bar on precision and text.


TOGETHER WITH HARDWIRE

The Future Just Got Physical

The Hardwire is a weekly newsletter tracking the next wave of AI-powered hardware: consumer devices, wearables, spatial computing, and the silicon shaping it all. From product launches to deep-tech inflection points, it connects what’s being built with why it matters.

Read by founders, engineers, investors, and operators who want to understand how AI escapes the cloud and enters the physical world.

If you care about where software meets atoms, The Hardwire is where the future gets wired.


NEW LAUNCH

Zoom Wants the Workflow, Not Just the Meeting

Zoom launched AI Companion 3.0, expanding from meeting summaries into agentic workflows that sit across meetings, docs, chat, and connected third party tools. A new web surface pulls context from conversations, files, and transcripts to generate tasks, drafts, and daily work views. It runs on a federated AI stack combining Zoom models with OpenAI, Anthropic, and open source options like NVIDIA Nemotron. The shift is about orchestration, not features.

Zoom is quietly aiming to become the default AI control layer for enterprise knowledge work. Meetings no longer end with notes. They auto-generate follow-ups, emails, documents, and tracked progress. This compresses entire categories like meeting assistants, lightweight PM tools, and AI note apps into a single surface IT already trusts. The real win is distribution. Zoom already sits inside daily work, so the AI ships straight into habit, not pilots.

Zoom finally realized summaries are table stakes. The real power is not better meetings, but owning what happens after them.


NEW TECH

AI Rewrites Cashmere’s Supply Chain

Everbloom built Braid.AI, a material science model that turns keratin waste into fibers that mimic cashmere and other textiles. Instead of breeding more goats, it tunes formulations and extrusion parameters to recreate softness, warmth, and feel using existing industrial spinning equipment. The bet is not a new fabric category, but a drop-in replacement that fits today’s mills, costs less, and scales without touching animal supply.

This is AI aimed squarely at industrial retrofitting, not green branding. Mills do not need new machines. Brands do not need new SKUs. Procurement can swap inputs without touching workflows. If it scales, animal fiber scarcity stops dictating pricing power, and sustainability shifts from marketing line item to default sourcing logic. The losers are specialty suppliers betting on scarcity. The winners are manufacturers who value interchangeability over provenance.

This is not climate tech theater. It is AI quietly moving upstream and pricing animal fibers out of relevance.


QUICK HITS

  • DoorDash has launched Zesty, an AI-powered social app that aggregates data across platforms to offer personalized restaurant discovery and sharing.
  • Meta has added conversation amplification to its AI glasses and a visual-triggered Spotify playback feature.
  • Adobe has upgraded Firefly with prompt-based video editing tools and expanded support for third-party image and video generation models.
  • Amazon is in talks to invest about $10 billion in OpenAI at a valuation potentially exceeding $500 billion.
  • Red Hat has acquired AI security startup Chatterbox Labs to strengthen security for models and MCP-based agents across its Red Hat AI platform.

TRENDING

Daily AI Launches

  • NexaSDK has launched an on-device AI SDK that runs multimodal models locally on iOS/Android with NPU acceleration and zero cloud costs.
  • Readever has launched a reading tool that allows you to chat with books and co-read with AI personas of historical figures.
  • Okara has launched a platform to run 30+ open-source models (like DeepSeek) with web search and file analysis, removing infrastructure hassles.
  • QualGent has launched an AI QA agent that automates mobile app testing on real devices using plain English prompts and self-healing scripts.
  • 🎥 KaraVideo unites all AI video models in one place.
  • 📚 Heardly is the Fast Way to read Best Book.
  • 🪶 CopyOwl is the First AI Research Agent, deep research on any topic in one click.
  • 🤖 Momen AI bridges the gap from vibe to viability, powering Lovable apps without code.

TOGETHER WITH US

AI Secret Media Group is the world’s #1 AI & Tech Newsletter Group, reaching over 2 million leaders across the global innovation ecosystem, from OpenAI, Anthropic, Google, and Microsoft to top AI labs, VCs, and fast-growing startups.

We operate the industry’s most influential portfolio of newsletters, each shaping a different frontier of the AI & Tech revolution:

Be Smarter in 5 Minutes

Discover the Future Products

We've helped promote over 500 Tech Brands. Will yours be the next?

Email our co-founder Mark directly at mark@aisecret.us if the button fails.