Goldman Sachs trading desk chief Rich Privorotsky believes Astra's capability leap could be the breakthrough AI bulls have been waiting for. Compared to mere price cuts, stronger model intelligence has the potential to unlock new application scenarios, broaden the AI demand curve, and keep the industry doubling down on computing power. This could reignite the AI capital expenditure cycle, with catalysts like Oracle's upcoming earnings and industry conferences in September also poised to add fuel.

OpenAI's release of the new Astra model has drawn widespread attention to the AI capability leap, and has revived the investment narrative around the AI spending cycle. In his latest market briefing, Goldman Sachs Delta One trading desk chief Rich Privorotsky said Astra looks "significant" — it has genuinely pulled ahead of its peers, and this may be exactly the kind of capability breakthrough AI bulls have been waiting for. This assessment quickly translated into market moves: SoftBank surged more than 10% overnight, and as a major OpenAI investor, its share price is viewed as a key barometer for shifts in OpenAI's valuation. Oracle also climbed roughly 5.5%. Yet in Privorotsky's view, the stock gains are the "least interesting part" of the story.

What deserves more attention is that OpenAI's internal assessment of Astra is equally aggressive. OpenAI President Greg Brockman called it a "generational leap" and suggested this moment may one day be seen as a landmark milestone on the road to AGI. At the end of the conference call, he declared directly: "Welcome to the AGI era." From Wall Street to inside OpenAI, Astra is being framed not merely as a model upgrade, but as a signal that a shift in AI capabilities could reshape the next capital expenditure cycle. This expectation has also driven the OpenAI ecosystem to clearly outperform the Anthropic camp this week.

Capability Breakthrough, Not Price Cuts: Reshaping the Demand Curve

In his briefing, Privorotsky drew a sharp distinction between two very different market signals: falling token prices versus a substantive leap in model capability. His core logic: when model prices drop, companies simply use the same amount of AI at a lower cost; but when models become "genuinely smarter," companies discover new application scenarios that were previously impossible, and the space for AI demand expands accordingly.

"A price decline is one thing; a true leap in intelligence will reshape the demand curve once again," Privorotsky wrote. "It forces all other labs to catch up, keeps the spending cycle alive, and revives the belief that there's something even better worth building." He also clarified that earlier commentary about "token demand growth not keeping pace with the rate of token cost decline" had been partially taken out of context. His complete reasoning: excess computing capacity only becomes a problem if model capabilities fail to make a fundamental leap — and the arrival of Astra points in precisely the opposite direction.

Astra Targets AGI: AI Competition Shifts from Price Wars to an Intelligence War

Astra's performance on some AGI-related benchmark tests has given OpenAI the confidence to claim it has surpassed Anthropic on certain metrics. Although the industry has no broadly accepted definition of AGI, nor unified testing standards, and Greg Brockman noted that the final judgment should be left to users, his stance on the media call was already quite unambiguous. When asked whether Astra meets the definition of AGI, he replied: "I believe this is the model."

This statement has further reinforced how the market is interpreting the release: competition among leading model developers is increasingly shifting from price and efficiency comparisons to a head-to-head contest in intelligence capability. For AI bulls, this is the signal they have long awaited — if the leading edge can be demonstrated in model intelligence itself, then the durability of the AI capital expenditure cycle gains new support.

September's AI Calendar: A Dense Catalytic Window, Bullish Sentiment May Keep Building

Privorotsky noted that September's market calendar looks constructive for the AI sector, with multiple potential catalysts providing an ongoing channel for bullish sentiment. Oracle is set to report earnings next week, with the market focused on progress in its cloud infrastructure and AI-related businesses. At the same time, tech industry conference season will offer additional catalysts, helping Astra's positive impact spread further across the broader AI supply chain.

"The return of breakthrough intelligence is exactly what's needed to reignite both the willingness and the necessity for sustained spending," Privorotsky concluded. Even so, he also cautioned that energy price trends and next week's CPI data remain the main potential disruptions to current market sentiment, and deserve close attention.

Risk Warning and Disclaimer

Markets carry risk, and investment requires caution. This article does not constitute personal investment advice, nor does it account for any individual investor's specific investment objectives, financial situation, or needs. Users should consider whether any opinions, viewpoints, or conclusions expressed herein align with their particular circumstances. Any investment decisions made based on this content are taken at your own risk.