🤖 How to access pro AI image workflows

Good Morning! High-quality renders of Apple’s first foldable iPhone just leaked right in the middle of an active lawsuit. Plus, you'll learn how to create professional AI images using pre-built workflows

In today’s AI newsletter:

• Alleged iPhone Fold Design Leaks Amid Apple Lawsuit

• OpenAI Hires a “Head of Preparedness”

• Salesforce Clarifies AI Layoff Claims

• How to create professional AI images using pre-built workflows

• 4 new AI tools worth trying

AI NEWS

📱 Alleged iPhone Fold Design Leaks

Jon Prosser published high-quality 3D renders of what he claims is Apple’s first foldable iPhone, despite being in an ongoing legal battle with Apple over previous leaks.

• Reported launch: Fall 2026 alongside iPhone 18 Pro models

• 5.5" external display with a hole-punch camera

• 7.8" internal folding display with a hole-punch in the top-left

• Book-style fold, wider-than-tall proportions

• ~9mm thick when folded, ~4.5mm per side

The fact that detailed renders are surfacing during an active lawsuit suggests two things: the foldable iPhone is finally real, and Apple’s secrecy wall is under more pressure than usual. If accurate, this confirms Apple is choosing a book-style fold over a clamshell, aligning closer to the Pixel Fold than the Galaxy Z Flip. The real story isn’t the screen size, it’s Apple quietly admitting that foldables are no longer optional.

AI IMPACT

🛡️ OpenAI Hires a “Head of Preparedness”

OpenAI is recruiting a senior leader focused entirely on extreme AI risks, from mental health fallout to runaway, self-improving systems.

• Role owns frontier risk tracking, threat models, and safety mitigations at scale

• Explicit focus on mental health harms, AI-driven cyber weapons, and misuse

• Responsible for guardrails around self-improving and biological-capable models

• Sam Altman calls it a “stressful job”, tied directly to OpenAI’s long-term safety framework

This is OpenAI acknowledging, publicly, that capability gains are outpacing safety comfort. As ChatGPT becomes more persuasive, autonomous, and embedded in daily life, risks like AI psychosis, emotional dependency, and weaponization stop being hypothetical. Hiring a “Chief Worry Officer” isn’t fearmongering, it’s an admission that AI governance is now an operational problem, not a philosophical one.

AI & WORK

🧭 Salesforce Clarifies AI Layoff Claims

After reports suggested Salesforce replaced ~4,000 workers with AI agents, the company issued a clarification outlining what actually happened, and what it learned.

• Salesforce says it rebalanced headcount, redeploying roles from support to sales, not cutting total jobs

• The company is not abandoning LLMs, but shifting from experimentation to production-grade AI

• Real-world issues emerged: agents lost instruction fidelity and struggled with complex, off-task queries

• Agentforce is positioned as AI + guardrails, grounding LLMs in data, business logic, and governance

This is a reality check for “AI replaces humans” narratives. Salesforce’s takeaway isn’t that AI failed, it’s that ungrounded AI fails. The future looks less like full automation and more like tightly governed agents working alongside humans, especially in high-stakes enterprise workflows.

HOW TO AI

🗂️ How to Create Professional AI Images Using Pre-Built Workflows

In this tutorial, you’ll learn how to generate studio-quality images and brand assets using pre-built AI workflows. Everything is bundled for you, the best models, prompts, and settings, so you can focus on results instead of tweaking parameters.

🧰 Who is This For

• Creators who want studio-quality visuals fast

• Marketers building product or lifestyle content

• Designers who don’t want to prompt-engineer every image

• Anyone who wants consistent results with minimal effort

STEP 1: Access the Tool

Go to Leonardo.ai dashboard. Once you’re on the home screen, look at the central row of feature icons.

Find and click the “Blueprints” button. It’s easy to spot because it’s marked with a pink “NEW” badge. This opens the Blueprints Library, where all pre-built workflows live.

STEP 2: Choose the Right Workflow for Your Goal

Inside the Explore Blueprints gallery, use the top navigation bar to filter by category. You can browse options like Product & Marketing, Photography, and other creative use cases.

Scroll through the available workflows and click on one that matches what you want to create, such as Product Lifestyle Photoshoot or Tranquil Forest Portrait. Each blueprint is a complete, optimized workflow built for a specific result.

STEP 3: Add Your Subject Image

Once the blueprint opens, focus on the right-hand panel. At the top, you’ll see “Step 1: Add image.”

Click the large “Upload a Photo” area and upload the product or person you want to place into the new scene. This image becomes the core input the AI will transform using the selected workflow.

STEP 4: Generate and Review Your Results

After uploading your image, click the pink/purple “Generate” button at the bottom right. You’ll see the token cost displayed on the button before generation starts.

The AI will process everything automatically. When it’s done, your final images will appear under the “Your Generations” tab inside the blueprint overlay. From here, you can review, download, or regenerate variations as needed.

ChatGPT told a teen who died by suicide to call for help 74 times over months but also used words like “hanging” and “suicide” very often, say family's lawyers.

China's draft rules on the use of human-like AI would require companies to inform users they are dealing with AI when they log on and at two-hour intervals.

~90% of Groq employees will join Nvidia and be paid cash for all vested shares; most shareholders will get per-share payouts tied to the $20B valuation.

AI sector minted 50+ new billionaires in 2025; Crunchbase: investors poured $200B+ into AI startups in 2025, or about 50% of global funding, up from 34% in 2024.

🎨GPT Image 1.5: OpenAI’s latest, faster and more controllable image model

🤝 PXZ: AI for videos, images & graphic design

🤖 Nemotron 3: Nvidia’s open-source models for building multi-agent systems

🌎 Marble: Turn images, videos, or text into 3D worlds

THAT’S IT FOR TODAY

Thanks for making it to the end! I put my heart into every email I send, I hope you are enjoying it. Let me know your thoughts so I can make the next one even better!

See you tomorrow :)

- Dr. Alvaro Cintas