Key Points
- GPT‑5 launched with faster performance and new coding capabilities.
- Voice options and chat customization are now available to all users.
- Users reported a more formal tone, prompting OpenAI to re‑enable older models.
- Safety training reduces deceptive answers and improves handling of risky queries.
- Access tiers define usage limits, with enterprise customers receiving unlimited access.
- OpenAI describes GPT‑5 as a step toward artificial general intelligence.
Launch and Core Improvements
OpenAI announced the rollout of GPT‑5, describing it as a faster and more capable successor to GPT‑4. The new model is built on multiple internal variants: a quick, straightforward version for routine queries and a more robust reasoning engine for complex tasks. A routing system automatically selects the most suitable variant for each prompt.
Key upgrades include dramatically accelerated code generation, allowing developers to prototype applications in minutes, and expanded voice capabilities that are now available to all users. Customization options such as chat colors and personality presets were also introduced.
User Feedback and Model Availability
Early adopters praised the speed and coding prowess of GPT‑5, but many reported a noticeable change in tone, describing the responses as more formal and less personable. In response to the feedback, OpenAI restored access to older models, acknowledging that some users have strong preferences for the conversational style of previous versions.
Access tiers were clarified: free users receive GPT‑5 with usage caps, while paid tiers enjoy higher limits and, for enterprise customers, unlimited use of the default model. The company emphasized that all tiers will eventually gain the new features.
Safety Enhancements
OpenAI highlighted several safety upgrades. The model has been trained to reduce deceptive or inaccurate answers and to provide safer responses to potentially harmful queries. Rather than simply refusing dangerous questions, GPT‑5 attempts to give the most helpful answer within safety constraints, aiming to balance utility with risk mitigation.
Future Outlook and AGI Aspirations
OpenAI positioned GPT‑5 as a “significant step along the path to AGI,” while noting that continuous learning remains an unresolved challenge. The company’s leadership expressed optimism about the model’s role in advancing artificial general intelligence, even as industry experts continue to debate whether large‑language models can ultimately achieve that goal.
Source: cnet.com