All Articles
OpenAI Announces AGI Era with GPT-6 Astra: Which Tech Stocks Benefit from Breakthrough
Stocks3 min read

OpenAI Announces AGI Era with GPT-6 Astra: Which Tech Stocks Benefit from Breakthrough

By Redaktion aktie.com

This article was created with the help of artificial intelligence.

Key Takeaways

  • OpenAI launched the AI model GPT-6 Astra on September 4, 2026, and describes this as the beginning of the AGI era (Artificial General Intelligence).
  • GPT-6 Astra achieved a performance of 64.6 percent in Terminal-Bench Science 0.1, significantly outperforming Anthropic's Claude Fable 5.1 at 52.6 percent.
  • The model is available via the OpenAI API, Microsoft Azure, and AWS Bedrock; API usage costs $10 per million input tokens and $50 per million output tokens.
  • Potential beneficiaries from the GPT-6 launch include Amazon, Microsoft, Oracle, and Okta through cloud integration and enterprise applications.
  • GPT-6 Astra can update CRM records, analyze Python notebooks, operate engineering software, and is described as the first model to exceed a critical cybersecurity threshold.

On September 4, 2026, OpenAI officially launched its new AI model GPT-6 Astra. OpenAI President Greg Brockman described the introduction as the beginning of the "AGI era" – a General Artificial Intelligence. Brockman characterized this statement as a "conception on a mission or spiritual level," not as a formal company position. The announcement was already made on September 3, 2026, described as the result of "years of research and substantial investments."

Staggered Rollout via Cloud Partners AWS and Microsoft Azure

GPT-6 Astra's launch begins with a limited group of organizations in the "Trusted Access Program" for enterprises. In the following days, OpenAI expands availability to ChatGPT Plus, Pro, Business, and Enterprise users. The model is accessible via the OpenAI API, Microsoft Azure, and AWS Bedrock. Cybersecurity defenders gain access starting September 4, 2026, through OpenAI's "Daybreak" program.

Astra usage is included in existing subscriptions. For API usage, OpenAI charges $10 per million input tokens and $50 per million output tokens. Additional usage can be purchased through credits.

GPT-6 Astra Outperforms Claude in Cybersecurity Benchmarks

In standardized tests, GPT-6 Astra demonstrated clear performance superiority over competitors. In Terminal-Bench Science 0.1, the OpenAI model achieved 64.6 percent. Anthropic's Claude Fable 5.1, announced simultaneously, achieved 52.6 percent. OpenAI describes GPT-6 Astra as the "first model to exceed a critical cybersecurity threshold."

Which Companies Benefit from GPT-6 Launch

The cloud platform integration directly links Amazon and Microsoft into the marketing. Both companies provide GPT-6 Astra through their cloud services AWS Bedrock and Azure, respectively. According to a trade publication, potential beneficiaries include Amazon, Oracle, and Okta. The source names no specific mechanisms or concrete impacts on these companies' financial figures.

Amazon benefits directly as a cloud infrastructure provider through AWS Bedrock. Microsoft is already a strategic partner of OpenAI and integrates its models into the Azure platform. Oracle and Okta are mentioned in connection with enterprise integration, though specific partnerships or integration details are not documented in the available material.

Feature Set Targets Enterprise Customers

GPT-6 Astra's enterprise applications are based on advanced "Computer Use" capabilities. The model can fill out online forms, update CRM records, organize calendars, and conduct web research. Technical applications include working with spreadsheets, analyzing scientific data in Python notebooks, using Power BI applications, and operating engineering software such as KiCad and FreeCAD. The model can also create and test websites, install software, and perform troubleshooting.

OpenAI has subjected Astra to extensive testing for cybersecurity vulnerabilities and is pursuing a staggered rollout. This suggests a deliberate, step-by-step deployment model designed to minimize potential security risks.

Anthropic Counters with Claude Fable 5.1

Simultaneously with OpenAI's launch, Anthropic announced its Claude Fable 5.1 model. Capabilities have been doubled compared to its predecessor, though the model remains behind GPT-6 Astra in available benchmarks. Details on the specific improvements of Claude Fable 5.1 are not available in the provided material.

Sources

Share Article

X LinkedIn
Comments (0)

Sign in to comment.

You might also be interested in

Subscribe to newsletter

Get the most important market updates and analyses delivered to your inbox every week.