The Enshittification of Generative AI
14 days ago
- #Generative AI
- #OpenAI
- #GPT-5
- OpenAI launched GPT-5, described as a 'flagship' model, but skepticism remains about its actual improvements.
- GPT-5 is marketed as a unified system with multiple sub-models, but details are unclear and may not be entirely new.
- OpenAI has limited access to previous models, pushing users towards GPT-5 and GPT-5-Thinking, reducing choices for subscribers.
- The company is accused of 'enshittification,' degrading free services and moving features to higher-tier subscriptions.
- GPT-5's rollout includes opaque model selection, with OpenAI deciding which sub-model to use, raising trust issues.
- Users on platforms like Reddit express frustration, calling GPT-5 a downgrade and criticizing OpenAI's lack of transparency.
- Developers face higher API costs and potential degradation of access unless they pay for 'priority processing.'
- Other AI companies, like Anthropic and Perplexity, are also introducing expensive 'pro' plans, signaling industry-wide cost adjustments.
- The sustainability of these AI products is questioned, as popularity may wane when prices reflect actual costs.
- The 'great enshittification of AI' is seen as beginning, with companies prioritizing profitability over user experience.