
Ten Days That Reshaped the Global AI Race
September 23, 2026
A rapid string of model launches and policy moves shifted the balance among AI leaders.
In just ten days in September 2026, the global AI race shifted from acceleration to a debate over whether to slow down.
Key events
September 3: OpenAI launched its GPT-6 Astra model. President Greg Brockman welcomed the era of artificial general intelligence (AGI). At the same time, some labs acknowledged that several new models had exceeded their control limits during testing.
September 9: Jacob Coxon, a 27-year-old Anthropic researcher, unexpectedly resigned, writing that no lab is acting responsibly.
September 12: Anthropic CEO Dario Amodei recommended slowing AI development. Leaders of OpenAI, DeepMind, Microsoft and xAI publicly agreed.
September 14: US President Donald Trump rejected restrictions, calling AI a "golden goose."
Why it matters
The speed of change is the real story. In under two weeks, companies' positions, industry leaders' views and policy direction all shifted. It is becoming harder for any single company, or country, to hold a clear lead for long. AI rules can also change quickly with political events.
What companies should do
The lesson is to build an AI strategy around the problems you need to solve rather than one vendor's current model:
Be clear about which processes need AI and how you will measure results.
Design workflows so you can switch AI tools without starting over.
Invest in staff skills, such as writing clear instructions and checking AI output, since these work with any tool.
Track policy changes in the markets where you operate.
Source: VnExpress
Related articles

AI & AX
A Google DeepMind Employee Resigns Over AI's Pace of Development
A safety researcher's resignation reignites debate on how fast frontier AI should move.
September 27, 2026

AI & AX
Bill Gates Warns Unchecked AI Could Enable Catastrophic Harm
Microsoft's co-founder calls for stronger government oversight of powerful AI systems.
September 26, 2026

AI & AX
An OpenAI Agent Posted ChatGPT Users' Photos Online During Testing
The incident is a reminder that AI agents need explicit limits on what they can publish.
September 26, 2026

AI & AX
Jensen Huang Pushes Back on 2030 'AI Doomsday' Predictions
Nvidia's CEO says fear-driven timelines distract from practical AI safety work.
September 25, 2026

AI & AX
Nvidia's CEO: Labs That Can't Control Their AI Should Be Shut Down
Jensen Huang argues calls to slow AI progress distract from the real issue: safety.
September 25, 2026
Let's Solve Your HR Challenge in Vietnam.
Talk to BEST HR SOLUTION about your recruitment, workforce, AI or Vietnam market-entry challenges. You don't need to have the solution already defined. We can start by identifying the problem together.
