Reports

OpenAI Launches API for gpt-image-1 Model

Hak

25 Apr 2025 — 3 min read

It's Friday, April 25, 2025, and you're reading the Agentive Daily Report.

Busy People‘s Section

OpenAI released its gpt-image-1 model via API, enabling developers to integrate the image generation technology at scale

Microsoft launched two specialized AI agents—Researcher and Analyst—as part of its vision for human-AI workplace teams

Microsoft's study of 31,000 workers revealed "Frontier Firms" that use AI agents are showing measurably higher productivity

Google's Gemini has reached 350 million monthly active users, though still trailing ChatGPT's reported 600 million

Perplexity AI alleges Google blocked its deal with Motorola to become the default AI assistant on their devices

MIT researchers created a "periodic table of machine learning" connecting 20+ classic algorithms to aid discovery of new techniques

Today's Top Stories

OpenAI Brings Powerful Image Generation to Developers via API

OpenAI has officially launched its gpt-image-1 model through API, making the same technology behind ChatGPT's viral image generation available to developers for integration into their own applications. With pricing at $0.02-$0.19 per image and flexible moderation controls, this move puts the company at the centre of the AI-powered creative ecosystem.

This is significant because it shifts the balance of image generation capabilities toward OpenAI's ecosystem, with major platforms like Adobe, Figma, and Canva already integrating the technology. By opening up API access to a model that generated over 700 million images in just one week after its March release, OpenAI ensures its technology becomes the default for design tools industry-wide.

Microsoft Identifies Rise of "Frontier Firms" Using AI Agents as Team Members

Microsoft's new Work Trend Index report, analyzing data from 31,000 workers across 31 countries, identifies the emergence of what it calls "Frontier Firms" - organizations powered by intelligence on tap and hybrid teams of humans and AI agents. The research shows 82% of leaders plan to deploy AI agents in the next 12-18 months, with nearly half already using them to fully automate workflows.

The implications are substantial: organisations implementing agent-human teams are showing higher productivity, while traditional org charts are giving way to "Work Charts" where teams form around goals rather than functions. Real-world examples show impressive results, with Wells Fargo cutting query time from 10 minutes to 30 seconds, and Bayer researchers each saving 6 hours weekly using agents.

MIT Creates "Periodic Table" for Machine Learning to Systematise AI Innovation

MIT researchers have introduced a "periodic table of machine learning," connecting over 20 classical algorithms in a framework that could reshape how AI systems are created. Rather than continuously reinventing techniques, this systematic approach enables researchers to see connections between existing methods and identify gaps where new algorithms could emerge.

This innovation has immediate practical applications – by mixing two existing techniques, the team created a new image classifier that outperforms top models by 8%. The broader implication is the potential for more deliberate, systematic AI innovation rather than accidental discovery. It creates a pathway for more explainable AI by making it easier to understand why models work and how they achieve results.

Other Developments Worth Noting

Google Gemini User Numbers Revealed: Court documents showed that Google's Gemini has about 350 million monthly active users, though still trailing ChatGPT's reported 600 million, revealing the competitive landscape in conversational AI.
Perplexity Alleges Google Interference: Perplexity claimed in court that Google's contract with Motorola effectively blocked the implementation of Perplexity as the default AI assistant on Motorola devices, highlighting growing competitive tensions.
AI for Personal Connection: A new analysis of AI usage reveals that "therapy & companionship" has become the most common use case for AI, with people turning to AI assistants for emotional support and conversation ahead of traditional utility functions like search.
Sam Altman Steps Down from Oklo: OpenAI's CEO has stepped down as chair of nuclear energy startup Oklo to give the company more flexibility in exploring partnerships with other AI firms, suggesting interconnections between energy and AI sectors.
US Government Mandates AI in Education: The US government has ordered the integration of AI into K-12 education, establishing AI courses for students and creating a White House Task Force on AI Education, signaling a policy shift toward AI literacy.

New Tools Discovered

Spring.new: An AI agent that builds custom SaaS applications from simple prompts, enabling rapid prototyping of software ideas.
Character AI AvatarFX: An AI video model for creating lifelike chatbot avatars with enhanced expressiveness and visual appeal.
Peek: An AI personal finance coach that guides users through financial decisions by analysing spending patterns.
Suna: An open-source agent that executes real-world tasks through conversation by navigating websites, managing files, and running system commands.
CUA (Computer Use Agents): A system that helps AI agents control computers within secure virtual containers, running macOS or Linux environments at near-native speed.

Discover more tools at Agentive.Directory

That's a wrap for today! Thank you for reading this report.

Have thoughts on today's edition? Hit reply and let us know what you're thinking. Or if you've discovered a cool AI tool we should feature, drop us a line.

Until tomorrow,
Hak from Agentive.Studio