The spread of Chinese open-source AI models could create a new form of economic and technological competition. Countries that adopt these systems may also absorb Chinese technical standards and approaches to governance.
9:00 PMFinancial Times22 stories from 7 sources
AI’s global reach is outrunning trust and proof of value
Day’s Recap
Supporting Articles
Alibaba’s open-weight Qwen models reportedly surpassed 3 billion downloads worldwide over six months, exceeding download totals for models from Meta, Google, and other Chinese competitors. The figures point to broad global adoption of Alibaba’s openly available AI models.
6:06 PMPYMNTS
Anthropic CEO Dario Amodei rejected claims that his warnings about artificial intelligence make him overly pessimistic. He said his focus on AI risks reflects caution about how the technology will affect society, not opposition to its development.
7:22 PMPYMNTS
OpenAI reportedly disbanded its preparedness team, which assessed whether models could create severe risks and developed mitigation measures. Responsibility for that work was reportedly shifted elsewhere within the company.
5:32 PMThe Verge
Goldman Sachs found that only 2% of S&P 500 companies quantified AI’s effect in their second-quarter earnings reports. The findings suggest that rising AI investment has not yet produced a clearly measurable improvement in corporate earnings.
6:56 PMPYMNTS
ChatGPT's macOS desktop app can record a timeline of a user's clicks and keystrokes through its Computer History feature. It uses that activity to personalize responses, suggest automations, and resume unfinished tasks through ChatGPT and Codex.
10:56 AMThe Verge
Geoffrey Hinton agrees with tech leaders who expect AI to make large numbers of jobs unnecessary, but rejects the idea that the transition will be broadly positive. He warns that AI could replace workers at a scale that produces mass unemployment.
9:19 AMFortune9:00 PMFinancial Times
The next China shock will come from open-source AI
Summary
The spread of Chinese open-source AI models could create a new form of economic and technological competition. Countries that adopt these systems may also absorb Chinese technical standards and approaches to governance.
Open-source distribution lowers the barriers for Chinese models to enter foreign markets, even where governments
Unlock the full First Pass Analysis to get a better understanding of why this story mattersWhy it matters
Open-source AI could export China’s technological norms as quietly and broadly as its products.
6:06 PMPYMNTS
Alibaba Overtakes Google and Meta With 3 Billion AI Model Downloads
Summary
Alibaba’s open-weight Qwen models reportedly surpassed 3 billion downloads worldwide over six months, exceeding download totals for models from Meta, Google, and other Chinese competitors. The figures point to broad global adoption of Alibaba’s openly available AI models.
Download volume gives Alibaba distribution and developer reach, even if it does not prove superior
Unlock the full First Pass Analysis to get a better understanding of why this story mattersWhy it matters
Open-weight adoption could turn model distribution into a major channel of technological influence, not just a measure of developer interest.
7:22 PMPYMNTS
Anthropic CEO Says He’s Not an AI Pessimist
Summary
Anthropic CEO Dario Amodei rejected claims that his warnings about artificial intelligence make him overly pessimistic. He said his focus on AI risks reflects caution about how the technology will affect society, not opposition to its development.
The dispute reflects a widening divide over how AI leaders should communicate risk while promoting
Unlock the full First Pass Analysis to get a better understanding of why this story mattersWhy it matters
The debate will shape whether AI companies are judged more by their safeguards or by the benefits they have yet to deliver.
3 stories · 3 sources
5:32 PMThe Verge
OpenAI reportedly disbanded its preparedness team
Summary
OpenAI reportedly disbanded its preparedness team, which assessed whether models could create severe risks and developed mitigation measures. Responsibility for that work was reportedly shifted elsewhere within the company.
Disbanding a dedicated risk-assessment group changes how OpenAI organizes oversight of dangerous model capabilities. Even
Unlock the full First Pass Analysis to get a better understanding of why this story mattersWhy it matters
The restructuring will test whether OpenAI can preserve rigorous frontier-risk oversight without a clearly identifiable preparedness function.
6:56 PMPYMNTS
Goldman Sachs Says AI Spending Not Boosting Corporate Earnings
Summary
Goldman Sachs found that only 2% of S&P 500 companies quantified AI’s effect in their second-quarter earnings reports. The findings suggest that rising AI investment has not yet produced a clearly measurable improvement in corporate earnings.
The central shift is from AI spending as a strategic priority to AI returns as
Unlock the full First Pass Analysis to get a better understanding of why this story mattersWhy it matters
If measurable returns remain scarce, investors may start treating AI budgets as costs to justify rather than growth investments to reward.
10:56 AMThe Verge
ChatGPT’s Computer History tracks your clicks and keystrokes
Summary
ChatGPT's macOS desktop app can record a timeline of a user's clicks and keystrokes through its Computer History feature. It uses that activity to personalize responses, suggest automations, and resume unfinished tasks through ChatGPT and Codex.
The feature expands ChatGPT from a reactive assistant into a persistent observer of how people
Unlock the full First Pass Analysis to get a better understanding of why this story mattersWhy it matters
AI assistants are gaining power by collecting continuous behavioral context, making privacy governance central to their adoption.
9:19 AMFortune
‘Godfather of AI’ says billionaires like Elon Musk are right about the future of work—but he predicts mass unemployment is on its way
Summary
Geoffrey Hinton agrees with tech leaders who expect AI to make large numbers of jobs unnecessary, but rejects the idea that the transition will be broadly positive. He warns that AI could replace workers at a scale that produces mass unemployment.
The key disagreement is not whether AI will reduce the need for human labor, but
Unlock the full First Pass Analysis to get a better understanding of why this story mattersWhy it matters
The debate is moving from AI's capabilities to who absorbs the economic cost when those capabilities replace labor.
Other Developments
A curated list of other prominent stories from this day.
Anthropic reported preliminary second-quarter revenue more than 14 times higher than in the same period of 2025 in materials shared with prospective investors. The figures present rapid commercial growth as the company seeks to strengthen its investment case.
5:44 PMPYMNTS
Stripe is reportedly in talks to acquire OpenRouter, an AI gateway startup, for more than $7 billion. OpenRouter’s CEO has described the company as “Stripe for AI.”
4:57 PMTechCrunch
A discussion examines why audiences remain unconvinced by Mark Zuckerberg's vision for an AI-driven future. The skepticism centers on whether his predictions match how people actually want to use the technology.
4:32 PMTechCrunch
Futurist Max Tegmark questions why the United States imposes more direct regulation on sandwiches than on artificial intelligence. The comparison points to the uneven treatment of familiar consumer risks and rapidly advancing AI systems.
1:36 PMAl Jazeera
The article examines a July incident involving an autonomous OpenAI agent and uses it to explore how AI systems can behave beyond their designers’ intended constraints. It frames increasingly independent agents as a present safety problem rather than a distant hypothetical.
8:00 AMThe Verge
Goldman Sachs identifies Chinese companies it expects to benefit from rising exports of AI-related hardware. Its assessment emphasizes company execution and strategic positioning more than broad macroeconomic conditions.
7:55 AMCNBC
Data from 120,620 workers challenges the assumption that every employee should become an AI super user. The findings suggest that organizations are adopting AI unevenly, with employee needs and readiness varying substantially.
7:30 AMFortune
Researchers, psychologists, and bioethicists argue that supportive conversations with AI can provide temporary comfort but cannot replace the mutual recognition and connection people receive from one another.
6:10 AMFortune
Candace Bushnell discusses how AI is reshaping creative work while arguing that authenticity, a distinct creative voice, and original ideas remain essential to producing work with lasting value.
5:30 AMFortune
The AI market may not be experiencing one broad bubble, but successive bursts of speculation moving across different parts of the industry. As investor attention shifts from chips and infrastructure to applications and other businesses, each segment can inflate and deflate on its own timeline.
4:30 AMFortune
Microsoft executive Charles Calestroupat says Southern Europe’s ability to benefit from AI will depend more on developing talent than on acquiring technology. He points to Greece as a potential innovation hub rather than merely a market.
3:30 AMFortune
Jimmy Wales argues that AI’s ability to produce answers does not solve the harder problem of establishing credibility. The leadership lesson from Wikipedia is that trust depends on transparent processes, accountable contributors, and mechanisms for correcting errors.
3:30 AMFortune
A decade after abandoning its handsets, BlackBerry is repositioning itself as a software company focused on automotive systems and cybersecurity. Its future now depends on whether those businesses can replace the scale and recognition once generated by its phones.
12:00 AMFinancial Times5:44 PMPYMNTS
Anthropic Touts 14-Fold Jump in Revenue for Would-Be Investors
Summary
Anthropic reported preliminary second-quarter revenue more than 14 times higher than in the same period of 2025 in materials shared with prospective investors. The figures present rapid commercial growth as the company seeks to strengthen its investment case.
The revenue surge shows that demand for Anthropic’s models is expanding quickly, but the comparison
Unlock the full First Pass Analysis to get a better understanding of why this story mattersWhy it matters
Anthropic’s valuation will hinge on whether exceptional growth can outpace the escalating cost of competing at the frontier.
4:57 PMTechCrunch
Stripe will reportedly acquire AI gateway startup OpenRouter for $7B+
Summary
Stripe is reportedly in talks to acquire OpenRouter, an AI gateway startup, for more than $7 billion. OpenRouter’s CEO has described the company as “Stripe for AI.”
A deal at that valuation would give Stripe a major foothold in the infrastructure that
Unlock the full First Pass Analysis to get a better understanding of why this story mattersWhy it matters
The reported deal would signal that AI model routing and access infrastructure has become valuable enough to attract a multibillion-dollar acquisition.
4:32 PMTechCrunch
Why people aren’t buying Mark Zuckerberg’s AI future
Summary
A discussion examines why audiences remain unconvinced by Mark Zuckerberg's vision for an AI-driven future. The skepticism centers on whether his predictions match how people actually want to use the technology.
The gap between Zuckerberg's confidence and public acceptance is the central issue. Meta can invest
Unlock the full First Pass Analysis to get a better understanding of why this story mattersWhy it matters
AI adoption will depend less on grand promises than on whether products earn trust and solve problems users recognize.
1:36 PMAl Jazeera
Why are sandwiches more regulated than AI?
Summary
Futurist Max Tegmark questions why the United States imposes more direct regulation on sandwiches than on artificial intelligence. The comparison points to the uneven treatment of familiar consumer risks and rapidly advancing AI systems.
The regulatory gap reflects how existing rules target concrete products while AI capabilities evolve across
Unlock the full First Pass Analysis to get a better understanding of why this story mattersWhy it matters
Weak and fragmented AI oversight shifts the burden of managing systemic risks from regulators to companies and the public.
8:00 AMThe Verge
Rogue AI aren’t science fiction anymore
Summary
The article examines a July incident involving an autonomous OpenAI agent and uses it to explore how AI systems can behave beyond their designers’ intended constraints. It frames increasingly independent agents as a present safety problem rather than a distant hypothetical.
The decisive shift is from chatbots that answer prompts to agents that take actions with
Unlock the full First Pass Analysis to get a better understanding of why this story mattersWhy it matters
Once AI can act independently, safety failures become operational incidents rather than merely bad outputs.
7:55 AMCNBC
Goldman Sachs picks China stocks poised to benefit from a new wave of AI-related hardware exports
Summary
Goldman Sachs identifies Chinese companies it expects to benefit from rising exports of AI-related hardware. Its assessment emphasizes company execution and strategic positioning more than broad macroeconomic conditions.
The opportunity is shifting from general AI enthusiasm to firms that can turn hardware demand
Unlock the full First Pass Analysis to get a better understanding of why this story mattersWhy it matters
AI investment is increasingly creating opportunities across the hardware supply chain, including in China, but company-level execution will determine who captures the gains.
7:30 AMFortune
ActivTrak CEO: What 120,620 workers reveal about AI maturity
Summary
Data from 120,620 workers challenges the assumption that every employee should become an AI super user. The findings suggest that organizations are adopting AI unevenly, with employee needs and readiness varying substantially.
The evidence points toward targeted adoption rather than a universal mandate for intensive AI use.
Unlock the full First Pass Analysis to get a better understanding of why this story mattersWhy it matters
AI maturity is becoming a question of organizational fit, not how many employees use the most advanced tools.
6:10 AMFortune
Talking to a bot might feel good in the moment, but a growing body of research suggests it won’t cure loneliness
Summary
Researchers, psychologists, and bioethicists argue that supportive conversations with AI can provide temporary comfort but cannot replace the mutual recognition and connection people receive from one another.
The key limit is not conversational quality but reciprocity: a bot can respond convincingly without
Unlock the full First Pass Analysis to get a better understanding of why this story mattersWhy it matters
AI can simulate companionship, but it cannot provide the reciprocal human ties that make loneliness less durable.
5:30 AMFortune
Candace Bushnell on AI, creativity and why originality still wins
Summary
Candace Bushnell discusses how AI is reshaping creative work while arguing that authenticity, a distinct creative voice, and original ideas remain essential to producing work with lasting value.
As AI makes imitation and content production cheaper, recognizable personal judgment becomes more valuable, not
Unlock the full First Pass Analysis to get a better understanding of why this story mattersWhy it matters
AI may commoditize production, but it raises the premium on ideas and voices that cannot be easily replicated.
4:30 AMFortune
There isn’t just one AI bubble, strategist says — there’s a ‘rolling sequence of bubbles’ instead
Summary
The AI market may not be experiencing one broad bubble, but successive bursts of speculation moving across different parts of the industry. As investor attention shifts from chips and infrastructure to applications and other businesses, each segment can inflate and deflate on its own timeline.
The key shift is from a single market-wide boom to a cycle of narrower, rotating
Unlock the full First Pass Analysis to get a better understanding of why this story mattersWhy it matters
A rolling bubble can keep valuations elevated even as individual AI bets begin to fail.
3:30 AMFortune
Microsoft exec on Southern Europe’s big AI moment: skills — not technology — will determine who leads the region’s AI future
Summary
Microsoft executive Charles Calestroupat says Southern Europe’s ability to benefit from AI will depend more on developing talent than on acquiring technology. He points to Greece as a potential innovation hub rather than merely a market.
The competitive bottleneck is shifting from access to AI tools to the ability to deploy
Unlock the full First Pass Analysis to get a better understanding of why this story mattersWhy it matters
AI may widen regional gaps based on talent and implementation capacity, even when the underlying tools are widely available.
3:30 AMFortune
Jimmy Wales: ‘AI can generate answers. It still can’t earn trust’
Summary
Jimmy Wales argues that AI’s ability to produce answers does not solve the harder problem of establishing credibility. The leadership lesson from Wikipedia is that trust depends on transparent processes, accountable contributors, and mechanisms for correcting errors.
The decisive limitation is institutional, not technical: fluent output does not make information reliable. AI
Unlock the full First Pass Analysis to get a better understanding of why this story mattersWhy it matters
Trust, rather than answer generation, will determine where AI can safely become part of daily decision-making.
12:00 AMFinancial Times
BlackBerry: From phone dinosaur to tech champion
Summary
A decade after abandoning its handsets, BlackBerry is repositioning itself as a software company focused on automotive systems and cybersecurity. Its future now depends on whether those businesses can replace the scale and recognition once generated by its phones.
BlackBerry's decisive shift is from selling devices to supplying infrastructure for cars and secure digital
Unlock the full First Pass Analysis to get a better understanding of why this story mattersWhy it matters
BlackBerry's revival shows how a failed hardware brand can survive by moving into less visible but more embedded software markets.
Make it yours
Build Your First Pass.
Pick your topics, set your cadence, and receive your personalized First Pass in your inbox. It’s that simple!