Computer-use changes the job
A chatbot gives answers.
A computer-using agent can take actions.
That changes the management problem.
Don’t just ask, “Is it smart?”
Ask:
What can it touch?
What should it prove?
When does it stop?
Who approves the consequence?
AI strategy is turning into “what should run where?”
Cloud for hard reasoning.
Local for private files.
Cheap models for repetitive jobs.
Frontier models when they earn the cost.
The next business skill isn’t picking one winner. It’s knowing where each job belongs.
Built this private family vacation command center with the new ChatGPT Work before Europe.
Now @ChatGPTapp is helping four of us navigate airports, Rome, trains, tours and logistics in real time.
AI with context is a whole different animal.
Field test underway. Full build breakdown after Europe.
@OpenAI - Sam, DMs open 😂
Promised I'd come back with the full breakdown after Europe. We actually used the thing, learned a ton, rebuilt parts of it after coming home, and just open sourced v1.2.0. Follow-up here 👇
xitter.cf/code__alchemist/status…
Brian Kolb retweeted
Introducing ChatGPT Images, powered by our flagship new image generation model.
- Stronger instruction following
- Precise editing
- Detail preservation
- 4x faster than before
Rolling out today in ChatGPT for all users, and in the API as GPT Image 1.5.
Love customizing these!
🚨PromptShare🚨
🎄 Fun prompt idea with Banana Pro:
Turn yourself into a super cute 3D character inside a Christmas ornament ✨🧸 Made in Firefly Board
The result? Pure festive magic.
PROMPT
A transparent Christmas bauble hanging by a red ribbon. Inside, a tiny diorama of the person from the reference reimagined as a cute 3d chibi character. He works at a mini futuristic AI desk with three glowing holo-screens showing neural networks and code. Add tiny plants, a mini coffee cup, soft desk lighting, floating UI icons, and snow-glitter at the base. Warm magical Christmas glow, cinematic reflections on glass, cozy high-end diorama aesthetic.
Cinematic lighting, shallow depth of field, soft reflections on the glass, ultra-polished materials, high detail, festive Christmas atmosphere. Whimsical, premium, and heartwarming."
👉 Now it’s your turn — create your own holiday bauble!
Repurposed it a little!
🤖 Made with AI
Prompt: Advertising-style food photography of [subject 1] arranged on a rustic table, with [subject 2] floating above in mid-air and a splash of [subject 1] frozen in motion. High-resolution photo manipulation with selective soft focus, clean negative space for text, dramatic lighting, and crisp detail.
Brian Kolb retweeted
Never would’ve guessed lol
PROMPT SHARE for Nano Banana Pro:
Who are your celebrity doppelgangers? I'm flattered by mine.
Works best with Aspect Ratio 1:1
"Role: Act as a senior facial recognition analyst and casting director.
Input: Analyze the attached photo of the person.
Task 1: Deep Feature Extraction
Scan the user's face for distinct biometric markers. Analyze:
Bone Structure: Cheekbone height, jaw width, and facial symmetry.
Features: Eye shape/tilt, nose bridge/tip, and lip morphology.
Expression & Vibe: The "energy" of the face (e.g., intense, friendly, stoic).
Task 2: Triangulate Matches
Identify the Top 3 celebrity matches from your dataset.
Match A: The closest overall structural/biometric match.
Match B: A strong match based on specific standout features (e.g., identical eyes or jaw).
Match C: A "mood" or stylistic match (similar vibe or grooming style).
Task 3: Calculate Likeness Scores
Assign a percentage match (0-100%) to each of the three celebrities based on feature correlation.
Task 4: Generate 4-Panel Output
Create a single composite image with a 2x2 Grid Layout titled "TOP CELEBRITY MATCHES".
Layout Instructions:
Top Left (Panel 1): The Original User Image (Input). Label: "ORIGINAL".
Top Right (Panel 2): Celebrity Match A. Label: "[Name] - [XX]%".
Bottom Left (Panel 3): Celebrity Match B. Label: "[Name] - [XX]%".
Bottom Right (Panel 4): Celebrity Match C. Label: "[Name] - [XX]%".
Ensure the celebrity images are high-quality headshots that match the angle of the user's photo as closely as possible for an accurate comparison."
Works great!
Of course, I couldn't miss this trend. But with my own color code. Here's the prompt.♥️
Try with nano banana pro 🍌 @eachlabs
"A professional, high-resolution profile photo, maintaining the exact facial structure, identity, and key features of the person in the input image. The subject is framed from the chest up, with ample headroom. The person looks directly at the camera. They are styled for a professional photo studio shoot, wearing a premium smart casual blazer in a subtle charcoal gray. The background is a solid '#562226' neutral studio color. Shot from a high angle with bright and airy soft, diffused studio lighting, gently illuminating the face and creating a subtle catchlight in the eyes, conveying a sense of clarity. Captured on an 85mm f/1.8 lens with a shallow depth of field, exquisite focus on the eyes, and beautiful, soft bokeh. Observe crisp detail on the fabric texture of the blazer, individual strands of hair, and natural, realistic skin texture. The atmosphere exudes confidence, professionalism, and approachability. Clean and bright cinematic color grading with subtle warmth and balanced tones, ensuring a polished and contemporary feel."
Fun and great creative thinking!
Prompt on Nano Banana Pro : Create an image about "[FILM_OR_NOVEL]" retold through a series of Polaroid photos pinned to a cork board. Each photo captures a key moment, with simple captions below. Arrange the photos in a loosely chronological path across the board, using colored strings to connect events and characters. Light the scene warmly to evoke nostalgia. Include incidental details, coffee cup rings, paper clips, handwritten notes, for authenticity.
Mathematics
Having fun with Nano Banana in Firefly Boards!
Try it and share your results!
A 4×4 grid showing the evolution of [subject] through history. Each of the 16 squares depicts [subject] in a different era, reflecting the technology, materials, style, and aesthetic of that period.
Each square has era-accurate colors, materials, and background. Perfectly aligned 4×4 grid, clean borders, ultra-detailed, cohesive lighting.































