OpenAI’s viral ImageGen heads to the masses

Good morning, AI enthusiasts. OpenAI just handed developers the reins to its viral image model, opening up SOTA AI creativity to hundreds of apps and services overnight.

With low cost pricing and the facility to render accurate text, edit images, and create diverse visual styles now embeddable in any app, OpenAI is muscling into workflows once dominated by its creative-focused rivals.

In today’s AI rundown:

• OpenAI unlocks powerful image creation via API

• Microsoft’s latest AI agents, workplace AI research

• Prepare for meetings immediately with Claude

• Ex-staff, experts challenge OpenAI’s restructuring

• 4 latest AI tools & 4 job opportunities

LATEST DEVELOPMENTS

OPENAI

🎨 OpenAI unlocks powerful image creation via API

The Rundown: OpenAI just launched its advanced image generation model, gpt-image-1, to developers via API — bringing the viral success of ChatGPT’s image capabilities to third-party applications and platforms.

The main points:

• The gpt-image-1 model powers ChatGPT’s image generation feature, which produced over 700 million images in only one week after its launch in March.

• The model enables high-quality image creation with varied styles, accurate text rendering, enhanced image editing, and more.

• OpenAI revealed that major platforms, including Adobe, Figma, and Canva, are already integrating the technology for skilled design workflows.

• Developers may also control the moderation level to tailor generated content safety, with standard “auto” filtering or less restrictive “low” moderation.

• Pricing is structured per token usage, with text prompts ($5 / 1M), input images ($10 / 1M), and output images ($40 / 1M), or ≈2-19c per image based on quality.

Why it matters: After ChatGPT’s image generator went viral last month, OpenAI is positioning itself to power the subsequent wave of AI-enhanced creative tools — and there can be a TON of them. The API release also deepens the broader creative ecosystem’s reliance on OpenAI, a position previously held by other rivals.

TOGETHER WITH WORKOS

🔑 Scale your AI app for enterprise

The Rundown: AI tools are reshaping how teams work—but enterprise adoption takes greater than innovation. It demands security. WorkOS provides drop-in APIs to allow you to meet enterprise requirements fast—without months of custom development.

With WorkOS, you may:

• Quickly integrate SSO, SCIM, audit logs, and more

• Get Enterprise Ready in days, not months

• Join OpenAI, Perplexity, and Cursor in scaling faster

Make your app Enterprise Ready with WorkOS today.

MICROSOFT

🤖 Microsoft’s latest AI agents, workplace AI research

The Rundown: Microsoft just released two latest Copilot agents, Researcher and Analyst, alongside its 2025 Work Trend Index report — which maps out the rise of AI-centric, human-led “Frontier Firms” set to reshape the workplace.

The main points:

• Researcher and Analyst bring deep reasoning to M365 Copilot for complex research and data science tasks like forecasting.

• The agents are rolling out as a part of Copilot’s “Frontier” early access program, alongside updates that permit corporations construct autonomous multi-agent systems.

• Microsoft’s research across 31,000 staff shows corporations leading in AI adoption are seeing major results:

  • 71% report their company is prospering vs 37% globally

  • 55% say they’ll handle increased workloads vs 20% globally

  • Employees show higher optimism about profession opportunities

• Microsoft also believes that each worker will develop into an “agent boss,” with all corporations becoming AI-human “Frontier Firms” for operations in 2-5 years.

Why it matters: Microsoft’s agent release coincides nicely with how the corporate views the longer term of labor: one dominated by AI agents, but managed by humans. We’re at first stages of a whole shift in how work gets done — with human-agent teams becoming the norm and corporations moving to completely hybrid, AI-infused structures.

AI TRAINING

🤝 Prepare for meetings immediately with Claude

The Rundown: On this tutorial, you’ll learn the best way to use Claude’s integration with Calendar and Gmail to simply research meeting participants, analyze previous communications, and get detailed company information multi functional place.

Step-by-step:

1. Head over to Claude and click on the settings menu to toggle Gmail and Calendar search.

2. Ask Claude to ascertain your calendar and research participants through the use of a prompt like: “Check my calendar for Thursday and supply a temporary summary concerning the participants and company.”

3. Review past communications by asking: “Check my email for previous conversations with [name] or someone from [company].”

4. Request to recommend talking points based on the combined insights.

Note: Only connect your accounts in the event you feel comfortable sharing your information with Claude. Also, you’ll be able to toggle these features on or off at any time for privacy control.

PRESENTED BY QWIET

🛡️ Revolutionizing AppSec with agentic AI

The Rundown: Qwiet AI’s Multi-AI Agent Approach revolutionizes application security (AppSec) through the use of multiple AI agents to offer enhanced code security and significantly improve efficiency between security and developer teams.

On this webinar, you’ll discover:

• Why multiple AI agents outperform single-agent solutions

• How Agentic AI corrects code vulnerabilities even when standard AI copilots fail

• The critical partnership between humans and AI in effective AppSec

Learn more and experience smarter AppSec with Qwiet AI.

OPENAI VS. THE INDUSTRY

❌ Ex-staff, experts challenge OpenAI’s restructuring

The Rundown: Greater than 30 AI experts and ex-OpenAI staffers published an open letter urging the attorneys general of Delaware and California to dam OpenAI’s restructuring, warning it could undermine its original mission to learn humanity.

The main points:

• 9 former OpenAI employees joined notable figures like AI ‘godfather’ Geoffrey Hinton in calling to dam the startup’s transition from nonprofit to for-profit.

• They argue the move will remove vital nonprofit oversight and safeguards, and redirect AGI development from public profit to shareholder returns.

• OpenAI needs transition approval from each state AGs by year-end to secure a pending $40B SoftBank investment contingent on the restructuring.

• The letter follows an earlier motion by 12 former employees looking for to weigh in on Elon Musk’s lawsuit against the corporate and CEO Sam Altman.

Why it matters: With billions in investments hanging within the balance and each former employees and AI heavyweights lining up against the AI leader, the opposition continues to mount against OpenAI’s for-profit shift — but we’ve seen little movement to suggest the transition is in jeopardy, despite the red flags being raised.

QUICK HITS

🛠️ Trending AI Tools

• 🗣️ Dia 1.6B – Open-source text-to-speech AI with emotion controls and more

• 🎬 AvatarFX – Create long-form, coherent talking avatars from a reference photo

• 🤖 Kortix Suna – Open-source AI agent built to work like a human

• 📋 ShumerPrompt – Discover and share powerful AI prompts

💼 AI Job Opportunities

• 📱 Pika – Software Engineer

• 🤖 Snorkel – AI Solutions Engineer

• 🧾 Synthesia – Financial Accountant

• 💼 Glean – Strategic Finance Lead

📰 Every thing else in AI today

Perplexity released its Perplexity Assistant app on iOS, allowing users to take agentic actions, access web browsing, and more on mobile using voice commands.

ByteDance’s Dreamina launched Seedream 3.0, a brand new text-to-image model that ranks No. 2 on Artificial Evaluation’ Image Arena Leaderboard behind only GPT-4o.

OpenAI is reportedly forecasting sales of $125B in 2029 and $174B in 2030, powered by AI agents, “latest products,” and API and user growth.

NVIDIA released its NeMo microservices suite, allowing enterprises to simply construct AI agents with optimized company data flywheels for high-quality performance.

BMW announced plans to integrate Chinese startup DeepSeek’s AI models into its latest vehicles within the region starting later this yr.

Tempus AI is partnering with biotech giants AstraZeneca and Pathos to develop the industry’s largest multimodal foundation model for cancer treatment discovery.

COMMUNITY

🎥 Join our next live workshop

Join our next workshop this Friday, April twenty fifth, at 4 PM EST with Dr. Alvaro Cintas, The Rundown’s AI professor. By the tip of the workshop, you’ll be able to bring your most complex ideas to life using Canva’s powerful creative tools.

RSVP here. Not a member? Join The Rundown University on a 14-day free trial.

🤝 Share The Rundown, get rewards

We’ll all the time keep this text 100% free. To support our work, consider sharing The Rundown with your pals, and we’ll send you more free goodies.

See you soon,

Rowan, Joey, Zach, Alvaro, and Jason—The Rundown’s editorial team