QATAR PERSPECTIVE

Doha deploys historical distance. Al Jazeera opens by recalling that DeepSeek 'stunned the world' a year earlier with a low-cost model rivaling American leaders. The new V4-Pro 'beats all competing open-source models in mathematics and coding' and only yields to Google's closed Gemini 3.1-Pro model in general knowledge.

But the most revealing detail is an admission from DeepSeek itself, cited by Al Jazeera: V4-Pro's performance 'remains slightly behind' GPT-5.4 and Gemini 3.1-Pro, 'suggesting a development trajectory that lags approximately 3 to 6 months behind frontier models.' For a Gulf media outlet, this transparency signals confidence: DeepSeek is not claiming to have surpassed Americans; it is stating exactly where it stands.

Al Jazeera cites Marc Andreessen, the influential venture capitalist close to Trump, who had called the January 2025 R1 release a 'Sputnik moment for AI.' The reminder is not incidental: if R1 was Sputnik, V4 is Apollo. For the Gulf, which is investing heavily in AI (Saudi Arabia and the UAE dedicate tens of billions to their national AI strategies), the question is now: should we bet on closed American models or open Chinese models?