Archive — US — techcrunch
Google Gemini’s AI image model gets a ‘bananas’ upgrade
Google upgrades its Gemini chatbot with the Gemini 2.5 Flash Image model, enhancing AI image editing capabilities and aiming to compete with OpenAI's offerings. This update enables precise edits based on user requests, preserves consistency in details, and integrates better world knowledge, while incorporating safeguards to prevent misuse of the technology. Users can now utilize this tool seamlessly for various personal projects.
Key points
- Google launches Gemini 2.5 Flash Image for enhanced AI image editing.
- The update improves user control over edits and preserves image details.
- Gemini aims to compete with OpenAI's popular image tools.
- Better integration of world knowledge allows combining multiple references in prompts.
- User engagement with Gemini is lower than with OpenAI's ChatGPT.
- The model includes safeguards against creating harmful content.
Why it matters
- AI image models are a critical competitive area among major tech firms.
- OpenAI increased ChatGPT's user base significantly with its image capabilities.
- Meta is also investing in AI image technology by licensing Midjourney's models.
- User privacy and consent remain pivotal in the development of AI tools.
With the rollout of Gemini 2.5 Flash Image, Google aims to elevate its standing in the AI image editing space, addressing both user creativity and safety concerns while stepping up competition against OpenAI and other tech giants.