ChatGPT
ChatGPT is an advanced large language model developed by openai, capable of generating text, code, and multimodal content. It serves as a central interface for accessing various AI capabilities, including generative image and video creation, as well as direct control of external creative software.
Capabilities & Evolution
- Multimodal Generation: Generates text, code, and multimodal content, expanding beyond pure text interfaces.
- RLCD & Decision Making: Research by co-inventor diogo-almeida indicates a strategic shift toward calibrated decision-making via rlcd, moving away from purely human-preferred text generation.
- Post-Transformer Architectures: The broader AI ecosystem is exploring alternatives to Transformer-Alternatives to address computational inefficiencies, including State-Space-Models and Recurrent-AI.
- Open Source & Local Deployment: Concurrent evolution in open-source agents like Hermes-AI emphasizes local deployment and Bot-Mode productivity.
Recent Developments (2026)
- DevDay 2026: Highlights include the introduction of “Dots,” an always-on personal agent, and the release of the gpt-61 model.
- Adobe Photoshop Control: ChatGPT has demonstrated the ability to directly control and execute complex editing tasks within Adobe Photoshop. This includes precise manipulation of color, simulation of long exposure effects, and execution of complex cutouts. This capability builds upon earlier integrations with Adobe Lightroom.
- See detailed analysis: ChatGPT’s Advanced Photoshop Control: Color, Long Exposure, and Complex Cutouts
References
- Glyn Dewis. “ChatGPT Can Now ACTUALLY Use Photoshop, and It’s Impressive.” ChatGPT’s Advanced Photoshop Control: Color, Long Exposure, and Complex Cutouts.