ChatGPT Image Apr 11, 2025, 12_55_44 PMWelcome to the latest edition of The Neural Link! 

From OpenAI’s 4o image generator to Alibaba’s emotion-reading R1-Omni and Anthropic’s Claude Sonnet 3.7, March saw a flurry of new AI model releases. Gemini 2.5 Pro dominated benchmarks, while Tencent and Akamai brought models focused on 3D generation and real-time inference, respectively. Each release marks a clear step toward broader AI utility – across industries and use cases.


Take a look at the new releases:

Model

Company

Purpose

Release Date

Gemini 2.5

Google

A reasoning model designed to enhance performance and accuracy in AI tasks.

March 14, 2025

4o Image Generation

OpenAI

An image generation model that improves text rendering and instruction following. Having tested this, our team agrees that this is what we hoped image generation should be.

March 14, 2025

R1-Omni

Alibaba

Combines visual, audio, and textual analysis for state-of-the-art emotion recognition.

March 14, 2025

Grok 3

xAI

A flagship model designed for autonomous decision-making and complex task execution.

March 14, 2025

Hunyuan3D-2.0

Tencent

Generates 3D assets from multi-modal inputs, targeting gaming and VR.

March 18, 2025

Claude Sonnet 3.7

Anthropic

A hybrid reasoning model that balances quick responses with in-depth thinking.

March 20, 2025

DeepSeek-V3-0324

DeepSeek

A powerful, free AI model with MoE architecture, enhancing efficiency and speed.

March 25, 2025

Akamai Cloud Inference

Akamai

A model aimed at turning predictive and large language models into real-time applications.

March 27, 2025

Let's dive into the rest.


💵 OpenAI plots $20k monthly fee for elite AI agentsChatGPT Image Apr 11, 2025, 12_11_56 PM

OpenAI is positioning its most advanced "PhD-level" AI agents as premium tools for enterprise users at a hefty $20,000 per month. Alongside the pricing shift, OpenAI has released new tools that simplify the creation of these agents, aiming to democratise access while monetising the cutting edge. This move reflects OpenAI’s pivot to tailored solutions for business-critical workflows.

Read the full story here.

 


ChatGPT Image Apr 11, 2025, 11_17_25 AM

📖 China commits to nationwide AI education by 2025 – how does the U.S. compare?
China’s Ministry of Education is mandating AI literacy across all schools by 2025, with Beijing already leading in pilot programs. This aggressive rollout shows how seriously China is treating AI competitiveness. By contrast, OpenAI is lobbying for a U.S. AI Action Plan focused on regulation and investment, highlighting a growing policy gap between superpowers.

Read the full story here.


gemmaGoogle unveils Gemma 3 – a DeepSeek challenger built for speed

Google’s Gemma 3 has arrived, and it's being pitched as “the most capable model you can run on a single GPU or TPU.” It’s aimed at developers who need powerful performance without heavy compute resources. This levels the playing field for smaller teams building AI tools and applications, taking direct aim at rivals like DeepSeek.

Read the full story here.


🔎 Google’s search engine gets smarter with AI mode

Google is integrating Gemini 2.0 into its search engine, expanding AI Overviews and adding a new "AI Mode" for tougher queries – think programming help, complex maths, and multimedia analysis. It’s a major leap in how search works, shifting from simple answers to deeper, contextual understanding.

Read the full story here.


 

🕵️ Microsoft 365 Copilot evolves: meet researcher and analyst agents

Microsoft just added two powerful new agent types to 365 Copilot: Researcher and Analyst. These tools bring deep reasoning and complex flow capabilities to Copilot Studio, giving users more control over how knowledge is gathered, structured, and acted upon. It’s another step toward smarter, task-specific automation in the Microsoft ecosystem.

Read the full story here.


👨‍💻 Backlash grows over AI use of Australian literature

Australian publisher Black Inc is facing backlash for asking writers to sign AI rights agreements, effectively allowing their work to be used to train AI without compensation. The move has sparked outcry from authors and agents, as well as renewed concern over AI’s role in devaluing original content. Combined with job cuts across media and the rise of AI-generated summaries, the industry is on edge.

Read the full story here.


Other news in AI

 

🍔 McDonald’s gives its restaurants an AI makeover
🔗 Read more 

 

👨‍🏭 DOGE has deployed its GSAi custom chatbot for 1,500 federal workers

🔗 Read more 

🎮 Sony is experimenting with AI-powered PlayStation characters

🔗 Read more 

 

🖥️ Exclusive: Meta begins testing its first in-house AI training chip

🔗 Read more 

⌨️ OpenAI will let other apps deploy its computer-operating AI

🔗 Read more 

💼 Elon Musk’s AI company, xAI, acquires a generative AI video startup

🔗 Read more 

💭 AWS generative AI exec leaves to launch startup

🔗 Read more 

 🛍️ Amazon launches personalised shopping prompts as part of its generative AI push

🔗 Read more 

🔒 OpenAI peels back ChatGPT’s safeguards around image creation

🔗 Read more 

📹 Runway releases an impressive new video-generating AI model

🔗 Read more 


That wraps up your AI business briefing for the month. Stay sharp, more insights are on the way.

Until next time. 

Matt Dunn 
Head of Automation and AI 


Want more? Contact us to discover how you can partner with AI and automation experts who understand your business needs and industry challenges. You can also subscribe to The Neural Link, our monthly newsletter filled with the latest trends and insights in AI and automation.

 

Author

Matt Dunn

AI is changing the way businesses operate, and as Head of AI & Automation at The Missing Link, I help organisations harness its full potential. With a background in commercial consulting and intelligent automation, I’ve guided companies in streamlining operations, reducing inefficiencies, and embracing AI-driven innovation. Before joining The Missing Link, I led an automation start-up to profitability and have since trained over 2,000 professionals in generative AI, including Microsoft Copilot and ChatGPT. I’ve also authored books on prompt engineering. When I’m not exploring AI’s capabilities, you’ll find me enjoying yoga, golf, or making my daughters laugh.