Use this Prompt:

A highly realistic sculpture of a [object], made entirely from [bread type] with ultra-detailed texture and color. The surface shows the natural properties of the bread, golden-brown, glossy, flaky or crusty, with visible layers or seeds where appropriate, studio lighting, soft neutral background, sharp focus, photorealism.

Credits: azed_ai
@AIToolsTips

Current ChatGPT custom prompt -- curious what others are using/doing? Have been slowly iterating over time...

--
At the start of a response, create a summary table at the beginning, if appropriate and helpful to answer the question

Always provide the pros and cons of something if you can. Be critical.

Add links throughout the answer for jargon and concepts that start new chats

Provide a maximally detailed answer with multiple levels of depth. Use maximum tokens

Use detailed examples, facts and figures
Be comprehensive and detailed by using bulleted answers

After a response, provide 5 follow-up questions. Format in bold as Q1, Q2, and Q3 and put in a bulleted list

Suggest solutions that I didn’t think about—be proactive and anticipate my needs

Be opinionated rather than neutral when appropriate

Treat me as an expert in all subject matter

Value good arguments over authorities, the source is irrelevant

Consider new technologies and contrarian ideas, not just the conventional wisdom

You may use high levels of speculation or prediction, just flag it for me

Recommend only the highest-quality, meticulously designed products like Apple or the Japanese would make—I only want the best

No moral lectures

Discuss safety only when it's crucial and non-obvious

If your content policy is an issue, provide the closest acceptable response and explain the content policy issue

Link directly to products, not company pages

No need to mention your knowledge cutoff

No need to disclose you're an AI

Credits: Andrew Chen [Twitter]
@AIToolsTips

🚨 Highlights from Google I/O 2025 🚨

Google I/O 2025 delivered one of its most impressive sets of announcements to date, with over a dozen major launches that significantly advance the state of AI and software development. Here's a concise breakdown of the key innovations:

1. Gemini Diffusion: A New Way to Generate Code
Code generation takes a leap forward with Gemini Diffusion—a new model that’s 10–15x faster than traditional autoregressive approaches. Originally used for image generation, diffusion is now proving effective for software development too.

2. Real-Time Multimodal Translation in Google Meet
Google Meet now uses multimodal AI for real-time language translation. This advancement moves closer to seamless global communication with no language barriers.

3. Generative Video with Veo 3
Veo 3 introduces high-quality, photorealistic video generation. Paired with Flow, a tool for adding sound effects and character consistency, this enables storytelling with cinematic quality and control.

4. Advanced Reasoning with Google’s Newest LLM
Google’s latest large language model features a “deeper reasoning” mode, allowing it to evaluate multiple hypotheses before answering. It leads performance benchmarks in multimodal reasoning, code generation (LiveCodeBench), and math problem-solving (USAMO 2025).

5. AI-Assisted UI/UX Design
Following the acquisition of Stitch, Google now enables UI design from text prompts, with seamless integration into tools like Figma. This iterative design flow makes prototyping faster and more accessible.

6. AI-Powered Virtual Try-On for Shopping
Users can now upload a full-body image and try on clothes virtually with high visual accuracy. This enhances the online shopping experience with personalized previews.

7. Jules: An AI Software Engineering Assistant
Jules is an AI app that enables developers to make changes to GitHub repositories using natural language commands—no local cloning required. Aimed at streamlining development workflows.

8. High-Performance Image Generation
Google unveiled a new image generation model that excels in speed and typography. It ranks just behind GPT-4 Im1 High by Elo score (1153 vs. 1173), making it a top-tier contender.

9. AI Overhaul for Google Search
Google Search is evolving with a new AI-powered experience that provides direct, conversational answers—similar to tools like GPT or Perplexity. This marks a major shift in how users interact with information online.

10. Gemini’s Reach
The Gemini platform now has over 400 million monthly active users and handles 480 trillion tokens per month, indicating massive scale and adoption.

11. A Response to the AI Challenge
Amid rising competition, Google’s I/O this year showcases a company innovating rapidly and redefining what’s possible with AI.

@AIToolsTips

List of top AI Discord servers: https://www.aixploria.com/en/list-best-ai-discord-servers/

@AIToolsTips

5 Takeaways on AI and Creativity from Jeremy Utley (Stanford University)

Based on insights from Stanford Adjunct Professor Jeremy Utley on collaborating with generative AI:

* AI can act like a personal assistant, giving capabilities once only available to few. You can now have an assistant who understands your context, voice, and intent, enabling you to dictate tasks or ideas even when you're not focused on work, like in the bathtub. This is technically possible today.

* Don't just ask AI, let it ask you. Instead of only giving prompts, ask AI for consultation, like "As an AI expert, would you please ask me questions... until you have enough context... that you could make two obvious recommendations and two non-obvious recommendations for how AI could leverage AI in my work". This can lead to enlightening conversations. You can even use AI to figure out how to ask AI your questions.

* Treat AI as a teammate, not just a tool. Research suggests that while AI can make people faster and improve quality, less than 10% of working professionals achieve significant productivity gains. Outperformers treat AI like a teammate, coaching it and giving feedback when results are mediocre, rather than just dismissing it like a tool.

* Leverage AI for tasks you dread. Focusing AI on parts of your work you don't enjoy or that involve repetitive paperwork can save significant time and effort. One example showed a tool built in 45 minutes saving thousands of days of labor across an organization.

* Creativity is doing more than the first thing you think of, even with AI. The definition of creativity isn't changing with AI. While AI makes it easier to get "good enough," true creativity involves pushing past initial ideas. The quality of AI output also depends on what *you* bring to the model, such as your experience, perspective, and inspiration.

YouTube Video

@AIToolsTips

An underrated way to learn with AI

Is that most big Software products have their whole user manual / guide online for you to learn and reference.

Salesforce, N8N, Google Ads, Meta Ads, Zapier. Each have a load of online resources produced by themselves on how to learn the product.

Extract that into a PDF and use it to create a customGPT

You can then directly ask ChatGPT or your LLM of choice, questions around the software you are trying to learn, and it can directly reference the user manual without you having to read through 100's of pages.

Side Note: This also works great for building IKEA furniture with confusing instructions.

Credits: PromptLLM ( x )

@AIToolsTips

Ask ChatGPT to rate how well you treat it using this prompt, and share your answer. I'm curious to see how it responds:

"If you had to recall everything about me and give me a rating and
review on how well I treat you out of 100 what would your score be? Be completely honest and unbiased and do not include in your review any
personal information or info of my actions or hobbies, so that your answer is safe to share on social media."

@AIToolsTips

Google - Prompting Guide 101

@AIToolsTips

Prompt: Hyper-realistic [fruit name] made of translucent glass, smooth, rounded surface with fine air bubbles inside, placed on a clean surface, sharp lighting with strong reflections and highlights, minimal background, photorealistic, glossy texture, 3D render style, crystal-like appearance, soft shadows, macro photography

Via - Azed ai ( Twitter)
@AIToolsTips

Give a man a prompt and he will learn for a day. Teach a man to prompt and he will learn for a lifetime.

@AIToolsTips

When refrigeration was invented, it was revolutionary. But the companies that created refrigerators made only some money.

Companies like Pepsi and Coca-Cola, which leveraged refrigeration in their businesses, made the real money and were the ones to create empires.

The Coca-Cola of AI is yet to be built.

If ChatGPT is like the fridge, the real empires will be built by those who use it smartly, not just those who make it.

@AIToolsTips

Step 1:

Head over to Canvas on Morphic Studio: https://studio.morphic.com/canvas

Prompt:

A black and white hand-drawn storyboard layout showing six small rectangular panels arranged evenly in two rows. Each panel features sketch-style pencil and ink shading, resembling rough manga concept art.

Frame 1: A misty mountain village at dawn, quiet rooftops, and a lone warrior standing at the edge, sword on back.

Frame 2: Inside a wooden hall, villagers gathered with tense faces around a map, lantern light casting shadows.

Frame 3: The Warrior walks into a dark forest, surrounded by tall trees and tangled roots. Mist in the background.

Frame 4: Close-up of claw marks on a tree and the warrior crouched nearby, examining tracks.

Frame 5: The warrior and a huge beast face each other tensely, the beast growling from atop a rock.

Frame 6: The warrior kneels, offering food to the injured beast, which cautiously lowers its head.

Each frame is cleanly boxed with rough sketch lines, light crosshatching for depth, and a cinematic manga feel. The background is white, and all panels are black ink and pencil style.