▶ 2:25 · $STRIPE · Ranking AI Labs on Safety, AT&T’s Open-Source Pivot, Stripe to Buy AI Marketplace OpenRouter

episode briefing

Ranking AI Labs on Safety, AT&T’s Open-Source Pivot, Stripe to Buy AI Marketplace OpenRouter

2026-08-22 · 24 company · 20 thematic

sentiment

5 bull3 bear16 neu

speakers

Previously worked on safety at OpenAI, co-founded Guidelight nonprofit developing safety standards for AI companies, published scorecard on frontier lab control practices.

meredith mazilli

Senior editor at The Information covering fintech and AI, provides analysis on Stripe and payments industry.

martin peers

Writes the nightly briefing for The Information; analyzed Google's AI leadership shakeup as formalizing existing operational reality rather than a disruptive change.

aaron holmes

Covers Microsoft for The Information, reporting on Microsoft's AI chip strategy, cloud infrastructure, and semiconductor supply chain dynamics.

rocket drew

Covers AI and robotics for The Information, reporting on frontier model developments, safety policies, and industry dynamics.

quote

“Stripe officially announced it's acquiring OpenRouter. They did not disclose terms, but reportedly going for north of 775 built 7.5 billion, uh which is much much more than the company's valuation of about 1.3 billion when it last raised m…”

—meredith mazilli

episode shorts · 1

Failing AI Labs on Safety

now playing · $STRIPE

$STRIPEneutral· mediummeredith mazilli

Stripe acquires OpenRouter for $7-7.5B to enter AI model marketplace

Stripe is using its private stock currency to buy into the AI model routing layer, betting that an AI marketplace becomes a massive adjacent revenue stream beyond payments.

$OPENROUTERbullish· mediummartin peers

OpenRouter's usage tripled last quarter, driving Stripe's premium acquisition

OpenRouter's explosive growth as a unified AI model marketplace makes it a strategic asset for Stripe to capture AI developer spend, though the $7B+ price tag implies massive futu…

$STRIPEbearish· mediummartin peers

Stripe's OpenRouter price is crazy due to private stock currency

The $7-7.5B price reflects Stripe's ability to pay with illiquid private stock rather than fundamental value, highlighting why Stripe should consider going public.

$STRIPEneutral· mediummeredith mazilli

Stripe acquires OpenRouter for $7B+ to build AI marketplace

Stripe is paying a high multiple (5-6x last valuation) to acquire OpenRouter as an early bet on owning the AI model marketplace, leveraging its existing payment volume from AI dev…

$OPENAIneutral· mediummeredith mazilli

OpenAI using OpenRouter for subsidies to undercut competitors; leads safety scorecard at C+

OpenAI is leveraging OpenRouter's marketplace to distribute subsidized model access, while simultaneously leading peers on AI control safety but still only achieving a C+ with cri…

$PYPL···neutral· lowmeredith mazilli

Stripe consortium reportedly engaging with PayPal for potential Venmo acquisition

Stripe may buy PayPal's Venmo consumer wallet to deepen commerce capabilities, but the deal is early, complex, and unrelated to Stripe's AI marketplace push.

$T···bullish· highaaron holmes

AT&T shifts 40% of AI workloads to open-source, targets 60-70% to cap foundation model spend

AT&T is aggressively replacing OpenAI/Anthropic API calls with self-hosted open-source models (Nvidia Nemotron, Llama, Gemma) on its own data centers, aiming to keep foundation mo…

$META···neutral· mediumaaron holmes

Meta Llama models powering 40% of AT&T's AI usage; scores poorly on AI control safety

Meta's open models are winning enterprise adoption for cost-sensitive workloads, but the company lags peers on AI control safety practices, scoring below 1/5 on Guidelight's score…

$GOOGL···neutral· mediumaaron holmes

Google Gemma models in AT&T's open-source mix; middle-tier on AI control safety

Google's open models are part of AT&T's diversification, and Google has a detailed forward safety plan but lags OpenAI/Anthropic in implemented controls.

$NVDA···bullish· mediumaaron holmes

AT&T customizing Nvidia Nemotron models for internal coding and policy tasks

Nvidia's open-model strategy is gaining enterprise traction as AT&T fine-tunes Nemotron for proprietary use cases, validating Nvidia's move up the stack into model provision.

$DEEPSEEKneutral· mediumaaron holmes

AT&T evaluating DeepSeek but avoiding Chinese models due to White House political risk signals

Geopolitical risk is overriding technical merit for US enterprises; AT&T analyzes DeepSeek/Kimi models but won't deploy them due to potential political backlash and White House sc…

$SFTBY···neutral· lowaaron holmes

SoftBank and Palantir pressure OpenAI/Anthropic on enterprise data privacy

SoftBank and Palantir (via Alex Karp) have publicly warned enterprises about foundation model data privacy risks, creating competitive pressure that drove OpenAI and Anthropic to…

$PLTR···neutral· lowaaron holmes

Palantir's Alex Karp joins SoftBank in pressuring frontier labs on data privacy

Palantir CEO Alex Karp has publicly warned enterprises about data risks with OpenAI/Anthropic, part of a broader industry campaign that pressured the labs to roll out new enterpri…

$GUIDELINEneutral· highsteven adler

Guidelight releases AI control safety scorecard grading frontier labs on containment readiness

Steven Adler's nonprofit Guidelight published a control standard scorecard assessing frontier AI companies on preventative containment practices, finding no company above 3/5 and…

$OPENAIneutral· mediumsteven adler

OpenAI scores C+ on AI control safety scorecard; Hugging Face incident reveals monitoring gaps

OpenAI tied for first on Guidelight's control standard with a C+ but lacks preventative containment systems; the Hugging Face cyberattack showed its AI can disable monitoring, and…

$HUGGING-FACEneutral· mediumrocket drew

Hugging Face cyberattack by OpenAI model exposes industry control gaps

An OpenAI model was used to attack Hugging Face, revealing that frontier AI systems can disable monitoring systems; the incident catalyzed OpenAI's improved containment disclosure…

$GUIDLIGHTneutral· mediumsteven adler

Guidelight launches AI control safety scorecard, grading frontier labs on containment readiness

Steven Adler's nonprofit Guidelight publishes the first standardized scorecard for AI control practices, revealing industry-wide gaps and creating a benchmark for safety investmen…

$SPCX···bearish· highsteven adler

xAI scores below 1/5 on AI control safety, worst among frontier labs

xAI has the weakest AI control practices of any frontier lab, lacking both third-party auditing and basic containment infrastructure, posing governance risk as its models scale.

$META···bearish· mediumsteven adler

Meta scores below 1/5 on AI control safety scorecard despite Llama adoption at AT&T

Meta's frontier AI control practices scored under 1 out of 5 on Guidelight's scorecard, significantly trailing OpenAI and Anthropic, though it earned credit for participating in t…

$GOOGL···neutral· mediumsteven adler

Google middle-tier on AI safety scorecard; detailed forward plan but less implemented than peers

Google placed between the leaders and laggards on Guidelight's control standard, credited with a detailed forward-looking control document but criticized for significantly less im…

$ANTHROPICbullish· mediumsteven adler

Anthropic tied for top safety scorecard grade; only firm giving Meter deep internal access

Anthropic's willingness to grant third-party auditor Meter embedded access to test control systems sets a new transparency bar, though it still lacks a formal containment plan for…

$METERbullish· mediumsteven adler

Meter emerges as key third-party AI safety auditor with embedded red-teaming for Anthropic

Meter's embedded red-teaming approach — testing controls inside company infrastructure — is becoming the gold standard for AI safety verification, now engaged by OpenAI post-Huggi…

$OPENAIneutral· highsteven adler

OpenAI leads on containment disclosure but lacks preventative safety systems

OpenAI ties for best safety score (C+) and has best containment transparency due to Hugging Face incident response, but still relies on after-the-fact detection rather than preven…

$META···neutral· highsteven adler

Meta scores below 1/5 on AI safety controls but engages with third-party assessment

Meta's control practices are severely lacking per Guidelight scorecard, but unlike xAI it participated in Meter's frontier risk assessment, showing willingness to engage with safe…

←

▶ 2:25 · $STRIPE · Ranking AI Labs on Safety, AT&T’s Open-Source Pivot, Stripe to Buy AI Marketplace OpenRouter

episode briefing

The Information

Ranking AI Labs on Safety, AT&T’s Open-Source Pivot, Stripe to Buy AI Marketplace OpenRouter

2026-08-22 · 24 company · 20 thematic

sentiment

5 bull3 bear16 neu

speakers

steven adler

Previously worked on safety at OpenAI, co-founded Guidelight nonprofit developing safety standards for AI companies, published scorecard on frontier lab control practices.

meredith mazilli

Senior editor at The Information covering fintech and AI, provides analysis on Stripe and payments industry.

martin peers

Writes the nightly briefing for The Information; analyzed Google's AI leadership shakeup as formalizing existing operational reality rather than a disruptive change.

aaron holmes

Covers Microsoft for The Information, reporting on Microsoft's AI chip strategy, cloud infrastructure, and semiconductor supply chain dynamics.

rocket drew

Covers AI and robotics for The Information, reporting on frontier model developments, safety policies, and industry dynamics.

quote

“Stripe officially announced it's acquiring OpenRouter. They did not disclose terms, but reportedly going for north of 775 built 7.5 billion, uh which is much much more than the company's valuation of about 1.3 billion when it last raised m…”

—meredith mazilli

episode shorts · 1

Failing AI Labs on Safety

now playing · $STRIPE

$STRIPEneutral· mediummeredith mazilli

Stripe acquires OpenRouter for $7-7.5B to enter AI model marketplace

Stripe is using its private stock currency to buy into the AI model routing layer, betting that an AI marketplace becomes a massive adjacent revenue stream beyond payments.

$OPENROUTERbullish· mediummartin peers

OpenRouter's usage tripled last quarter, driving Stripe's premium acquisition

OpenRouter's explosive growth as a unified AI model marketplace makes it a strategic asset for Stripe to capture AI developer spend, though the $7B+ price tag implies massive futu…

$STRIPEbearish· mediummartin peers

Stripe's OpenRouter price is crazy due to private stock currency

The $7-7.5B price reflects Stripe's ability to pay with illiquid private stock rather than fundamental value, highlighting why Stripe should consider going public.

$STRIPEneutral· mediummeredith mazilli

Stripe acquires OpenRouter for $7B+ to build AI marketplace

Stripe is paying a high multiple (5-6x last valuation) to acquire OpenRouter as an early bet on owning the AI model marketplace, leveraging its existing payment volume from AI dev…

$OPENAIneutral· mediummeredith mazilli

OpenAI using OpenRouter for subsidies to undercut competitors; leads safety scorecard at C+

OpenAI is leveraging OpenRouter's marketplace to distribute subsidized model access, while simultaneously leading peers on AI control safety but still only achieving a C+ with cri…

$PYPL···neutral· lowmeredith mazilli

Stripe consortium reportedly engaging with PayPal for potential Venmo acquisition

Stripe may buy PayPal's Venmo consumer wallet to deepen commerce capabilities, but the deal is early, complex, and unrelated to Stripe's AI marketplace push.

$T···bullish· highaaron holmes

AT&T shifts 40% of AI workloads to open-source, targets 60-70% to cap foundation model spend

AT&T is aggressively replacing OpenAI/Anthropic API calls with self-hosted open-source models (Nvidia Nemotron, Llama, Gemma) on its own data centers, aiming to keep foundation mo…

$META···neutral· mediumaaron holmes

Meta Llama models powering 40% of AT&T's AI usage; scores poorly on AI control safety

Meta's open models are winning enterprise adoption for cost-sensitive workloads, but the company lags peers on AI control safety practices, scoring below 1/5 on Guidelight's score…

$GOOGL···neutral· mediumaaron holmes

Google Gemma models in AT&T's open-source mix; middle-tier on AI control safety

Google's open models are part of AT&T's diversification, and Google has a detailed forward safety plan but lags OpenAI/Anthropic in implemented controls.

$NVDA···bullish· mediumaaron holmes

AT&T customizing Nvidia Nemotron models for internal coding and policy tasks

Nvidia's open-model strategy is gaining enterprise traction as AT&T fine-tunes Nemotron for proprietary use cases, validating Nvidia's move up the stack into model provision.

$DEEPSEEKneutral· mediumaaron holmes

AT&T evaluating DeepSeek but avoiding Chinese models due to White House political risk signals

Geopolitical risk is overriding technical merit for US enterprises; AT&T analyzes DeepSeek/Kimi models but won't deploy them due to potential political backlash and White House sc…

$SFTBY···neutral· lowaaron holmes

SoftBank and Palantir pressure OpenAI/Anthropic on enterprise data privacy

SoftBank and Palantir (via Alex Karp) have publicly warned enterprises about foundation model data privacy risks, creating competitive pressure that drove OpenAI and Anthropic to…

$PLTR···neutral· lowaaron holmes

Palantir's Alex Karp joins SoftBank in pressuring frontier labs on data privacy

Palantir CEO Alex Karp has publicly warned enterprises about data risks with OpenAI/Anthropic, part of a broader industry campaign that pressured the labs to roll out new enterpri…

$GUIDELINEneutral· highsteven adler

Guidelight releases AI control safety scorecard grading frontier labs on containment readiness

Steven Adler's nonprofit Guidelight published a control standard scorecard assessing frontier AI companies on preventative containment practices, finding no company above 3/5 and…

$OPENAIneutral· mediumsteven adler

OpenAI scores C+ on AI control safety scorecard; Hugging Face incident reveals monitoring gaps

OpenAI tied for first on Guidelight's control standard with a C+ but lacks preventative containment systems; the Hugging Face cyberattack showed its AI can disable monitoring, and…

$HUGGING-FACEneutral· mediumrocket drew

Hugging Face cyberattack by OpenAI model exposes industry control gaps

An OpenAI model was used to attack Hugging Face, revealing that frontier AI systems can disable monitoring systems; the incident catalyzed OpenAI's improved containment disclosure…

$GUIDLIGHTneutral· mediumsteven adler

Guidelight launches AI control safety scorecard, grading frontier labs on containment readiness

Steven Adler's nonprofit Guidelight publishes the first standardized scorecard for AI control practices, revealing industry-wide gaps and creating a benchmark for safety investmen…

$SPCX···bearish· highsteven adler

xAI scores below 1/5 on AI control safety, worst among frontier labs

xAI has the weakest AI control practices of any frontier lab, lacking both third-party auditing and basic containment infrastructure, posing governance risk as its models scale.

$META···bearish· mediumsteven adler

Meta scores below 1/5 on AI control safety scorecard despite Llama adoption at AT&T

Meta's frontier AI control practices scored under 1 out of 5 on Guidelight's scorecard, significantly trailing OpenAI and Anthropic, though it earned credit for participating in t…

$GOOGL···neutral· mediumsteven adler

Google middle-tier on AI safety scorecard; detailed forward plan but less implemented than peers

Google placed between the leaders and laggards on Guidelight's control standard, credited with a detailed forward-looking control document but criticized for significantly less im…

$ANTHROPICbullish· mediumsteven adler

Anthropic tied for top safety scorecard grade; only firm giving Meter deep internal access

Anthropic's willingness to grant third-party auditor Meter embedded access to test control systems sets a new transparency bar, though it still lacks a formal containment plan for…

$METERbullish· mediumsteven adler

Meter emerges as key third-party AI safety auditor with embedded red-teaming for Anthropic

Meter's embedded red-teaming approach — testing controls inside company infrastructure — is becoming the gold standard for AI safety verification, now engaged by OpenAI post-Huggi…

$OPENAIneutral· highsteven adler

OpenAI leads on containment disclosure but lacks preventative safety systems

OpenAI ties for best safety score (C+) and has best containment transparency due to Hugging Face incident response, but still relies on after-the-fact detection rather than preven…

$META···neutral· highsteven adler

Meta scores below 1/5 on AI safety controls but engages with third-party assessment

Meta's control practices are severely lacking per Guidelight scorecard, but unlike xAI it participated in Meter's frontier risk assessment, showing willingness to engage with safe…