OpenAI Declares 'AGI Era' with Release of GPT-6 Astra Model

OpenAI’s GPT-6 Astra is the first model rated as ‘critical’ under its safety framework, as announced by President Greg Brockman. This model excels in benchmarks for math, coding, and cybersecurity, indicating significant advancements in performance. Notably, GPT-6 Astra has also uncovered two previously unknown zero-day vulnerabilities, highlighting its capabilities in cybersecurity. Brockman stated that this release marks the start of the ‘AGI era.’ This follows OpenAI’s recent focus on enhancing cybersecurity features in its models, as seen with Astra’s predecessor. Practitioners can expect more robust safety measures and performance benchmarks in future AI applications. For further details, see The Decoder.