TL;DR:
- ChatGPT 4o offers businesses powerful image generation even as the copyright debate continues.
- Gemini 2.5 Pro’s reasoning-based approach revolutionises AI by deeply examining information across multiple formats.
- Apple Intelligence offers privacy and personalisation but fails to attract users who see it as more novelty than necessity.
Written by Drew Schmitz with help from Claude Sonnet 3.7
ChatGPT 4o: Image Generation
OpenAI’s ChatGPT 4o has arrived with improved image generation, which is attracting much attention. The launch saw a million new users flood the platform within just one hour, and for good reason.
The upgraded system creates clear text in images and follows visual instructions remarkably well. I’m not sure what your experience has been, but previously the image generator struggled significantly with text rendering, often producing garbled letters or completely unreadable words. This change makes AI image creation useful for real content creation, rather than just being a fun tool.
For businesses, the benefits are considerable. Marketing teams can quickly create campaign mock-ups, product teams can present concepts without needing designers, and content creators can produce custom visuals on demand, all at a much lower cost.
However, copyright concerns are growing as these AI systems can generate images that mimic specific art styles or even particular artists’ work. Companies must carefully monitor how they use this technology to avoid potential legal issues, as copyright laws vary significantly between countries and are still catching up to these new capabilities.
But human designers are still essential. While AI produces good visuals, professional designers offer creativity and understanding that AI lacks for now. The best approach is to use AI for efficiency, alongside human creative thinking.
Google Gemini 2.5 Pro
Google’s Gemini 2.5 Pro is excelling in industry tests for reasoning and coding. What sets it apart is how it works: instead of merely predicting text patterns, it deeply examines information, draws logical conclusions, and uses more sophisticated decision-making.
This is a big shift from how most AI models operate. Gemini’s success may prompt other AI developers to adopt similar reasoning-based approaches. The model can process information across text, code, images, and audio formats, making it more versatile than earlier AI systems.
For businesses, this provides a powerful new tool. Development teams can use their coding abilities to automate routine tasks and detect bugs more efficiently. Marketing teams can better analyse customer feedback, while product teams can more accurately predict market responses to new features.
The model’s reasoning capabilities enable it to solve complex problems that previously required human expertise. By processing large volumes of data and extracting meaningful patterns, Gemini 2.5 Pro helps businesses make more informed decisions. These enhancements allow organisations to tackle more demanding challenges with greater precision, helping them stay competitive in fast-moving markets.
What is Apple Doing in AI?
Apple has finally jumped into the AI race with Apple Intelligence, but is anyone actually using it? Despite the flashy promises of personalised, private AI experiences, the adoption seems lacklustre at best.
They’ve packed in all the expected features: ChatGPT-powered writing tools, an AI image playground, personalised “Genmoji” creation, and a revamped Siri with actual natural language capabilities. Yet at Intelligence Assist, all six of our Apple device users have completely ignored these features.
The problem? Apple Intelligence feels more like a tech demo than something that solves real problems. While Google and Microsoft develop AI that transforms productivity and creative workflows, Apple seems content creating digital toys that look impressive in keynotes but gather digital dust in real life.
The privacy-focused approach is commendable, but what good is the most private AI if it’s not useful enough to open? Apple needs to either innovate meaningfully or risk becoming the tech equivalent of showing up overdressed to a casual party, technically impressive but missing the point entirely.
Want to stay informed about AI advancements?
Sign up for Intelligence Assist’s monthly newsletter to keep pace with cutting-edge technology.
