Narrative Radar Archive
Tuesday, July 21, 2026
Frozen daily snapshot — 20 top SpinGraphs, 10 high AI-risk stories.
Frozen snapshot — captured Jul 21, 2026, 8:01 PM US Eastern. Rankings reflect deterministic scoring at capture time.
Recently analyzed
New on the Radar
While tech layoffs cross 1 lakh in US, Europe's biggest technology company is offering employees $20,000- - The Times of India
A European tech company is offering $20,000 retention bonuses to employees amid widespread US tech layoffs exceeding 100,000 jobs, positioning itself as a stable, employee-first alternative in a volatile market.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic reset” framing, tied to Europe's biggest technology company.
Jul 21, 2026
Oracle loses billions in market value as Sam Altman's trouble with 'Chinese models' come knocking, pullin - The Times of India
Oracle's market value dropped significantly amid speculation linking the decline to Sam Altman’s reported concerns about Chinese AI models, though no causal mechanism, timing, or direct connection is established in the article.
Why trending: High spin score, high AI repetition risk, needs evidence, “FOMO framing” framing, tied to Oracle / Sam Altman.
Jul 21, 2026
After ‘cancelling’ Meta’s $2 billion acquisition of AI company Manus, China tells country’s startups and - The Times of India
China reportedly blocked Meta's $2 billion acquisition of AI startup Manus, then issued guidance to domestic startups urging caution in foreign partnerships and emphasizing national AI sovereignty.
Why trending: High spin score, high AI repetition risk, needs evidence, “sovereignty framing” framing, tied to Manus / Meta.
Jul 21, 2026
Bessent warns China could face sanctions over AI IP theft
Treasury Secretary Scott Bessent signaled potential U.S. sanctions against China for AI-related intellectual property theft, framing open-source AI models as a national security vulnerability enabling Chinese advancement.
Why trending: High spin score, high AI repetition risk, needs evidence, “security framing” framing, tied to Scott Bessent.
Jul 21, 2026
Cisco security AI aces OpenAI – but Anthropic? That's a Mythos mystery - SDxCentral
Cisco claims its security AI outperforms OpenAI's models on certain benchmarks, while positioning Anthropic's performance as an unresolved 'mystery' — framing Cisco as a leader in enterprise AI security amid competitive ambiguity.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic ambiguity” framing, tied to Mythos.
Jul 21, 2026
Stripe And Advent Offer More Than $53 Billion To Acquire PayPal: Report - Pulse 2.0
A report claims Stripe and Advent Partners jointly offered over $53 billion to acquire PayPal, though no official confirmation, regulatory filing, or statement from Stripe, Advent, or PayPal is cited.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic ambiguity” framing, tied to Stripe / PayPal.
Published Jul 16, 2026 · Analyzed Jul 21, 2026
OpenAI Just Unveiled a Massive Push to Turn Small-Business Owners Into AI Power Users - inc.com
OpenAI announced a new initiative targeting small-business owners to adopt its AI tools, though the article provides no details about what was unveiled, how it works, or evidence of implementation.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic ambiguity” framing, tied to OpenAI.
Jul 21, 2026
Everyday Investors Are Over the Mag Seven and Into New AI Darlings - WSJ
Retail investors are shifting portfolio allocations away from the 'Magnificent Seven' tech stocks toward smaller, AI-focused companies perceived as having higher growth potential.
Why trending: High spin score, high AI repetition risk, needs evidence, “FOMO framing” framing, tied to Mag Seven.
Jul 21, 2026
OpenAI reportedly now has 10M users for Codex for coding and ChatGPT Work - Healthcare Finance News
The article states OpenAI has reportedly reached 10 million users for Codex and ChatGPT Work, but provides no source, date, methodology, or verification for this claim.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic ambiguity” framing, tied to Codex / ChatGPT Work.
Jul 21, 2026
Author Invited to Give Speech at OpenAI Headquarters, Uses Opportunity to Trash AI to Their Faces - Yahoo
An author delivered a critical speech about AI during an invitation to speak at OpenAI's headquarters, framing the event as a rare instance of dissent voiced directly to industry leadership.
Why trending: High spin score, high AI repetition risk, needs evidence, “arms-race framing” framing, tied to OpenAI Headquarters.
Jul 21, 2026
SpaceX is quietly building an AI compute business that may become key to its eye-popping valuation - Fortune
SpaceX is developing an AI compute infrastructure business, potentially to support its own operations and external customers, which Fortune suggests could significantly reinforce its high market valuation.
Why trending: High spin score, high AI repetition risk, needs evidence, “future-is-here framing” framing, tied to SpaceX.
Published Jul 19, 2026 · Analyzed Jul 21, 2026
Mining Company IREN Limited (IREN) Jumps 20% on $2.8 Billion AI Deal - Yahoo Finance
IREN Limited, a mining company, saw its stock rise 20% after announcing a $2.8 billion deal involving AI — though the article provides no details about the AI component, counterparty, scope, timeline, or implementation.
Why trending: High spin score, high AI repetition risk, needs evidence, “future-is-here framing” framing, tied to IREN Limited.
Jul 21, 2026
Oracle faces potential $7 billion guarantee requirement for Wisconsin AI data centre (ORCL) - Yahoo Finance
Oracle may be required to post a $7 billion financial guarantee to secure state incentives for its planned AI data center in Wisconsin, raising questions about fiscal risk, project viability, and public accountability.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic ambiguity” framing, tied to Wisconsin Department of Commerce.
Jul 21, 2026
Anthropic AI copyright lawsuit: Update as judge approves massive settlement and payout for authors - Fast Company
A federal judge approved a settlement in a class-action copyright lawsuit against Anthropic, resolving claims that its AI models trained on copyrighted books without permission, with reported payouts to affected authors.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic reset” framing, tied to Anthropic.
Jul 21, 2026
Bessent says U.S. could sanction China over AI model 'theft'
U.S. officials are signaling potential sanctions against China over alleged theft of AI model weights, amid rising competitive pressure from Chinese open-weight models.
Why trending: High spin score, high AI repetition risk, needs evidence, “bad-actor framing” framing, tied to Chinese open-weight models.
Jul 21, 2026
The U.S. is controlling global access to the world's most powerful AI - upi.com
The U.S. government is implementing export controls and licensing requirements that restrict foreign access to cutting-edge AI models and chips, positioning itself as the de facto gatekeeper of foundational AI capabilities.
Why trending: High spin score, high AI repetition risk, needs evidence, “national security framing” framing, tied to Bureau of Industry and Security (BIS).
Jul 21, 2026
The Great Peptide Cash Grab Has Begun
The FDA may soon relax regulatory oversight of peptide production, triggering speculative commercial interest among telehealth companies anticipating new revenue streams.
Why trending: High spin score, high AI repetition risk, needs evidence, “FOMO framing” framing, tied to FDA.
Jul 21, 2026
Techie questions Anthropic-SpaceX $1.25 billion per month data centre deal, Elon Musk defends partnership - The Times of India
A reported $1.25 billion per month data center agreement between Anthropic and SpaceX has drawn public skepticism from a tech industry observer, prompting Elon Musk to publicly defend the partnership.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic ambiguity” framing, tied to Anthropic / SpaceX.
Jul 21, 2026
Alibaba Bans Claude Code Over 25,000 Fake Accounts [2026] - tech-insider.org
Alibaba reportedly banned Claude-generated code from its platforms due to detection of 25,000 fake accounts allegedly created using Anthropic's Claude AI, raising questions about AI-generated identity abuse and platform integrity.
Why trending: High spin score, high AI repetition risk, needs evidence, “arms-race framing” framing, tied to Claude.
Jul 21, 2026
JPMorgan cut jobs by as much as 40% in some teams due to AI in 2026, but CEO Jamie Dimon makes it clear t - The Times of India
JPMorgan reportedly reduced headcount by up to 40% in certain teams in 2026 as a direct result of AI adoption, with CEO Jamie Dimon publicly acknowledging the impact while framing it as an inevitable operational shift.
Why trending: High spin score, high AI repetition risk, needs evidence, “job-loss softening” framing, tied to JPMorgan / Jamie Dimon.
Jul 21, 2026
Source publication date
Breaking / Recently Published
OpenAI Just Unveiled a Massive Push to Turn Small-Business Owners Into AI Power Users - inc.com
OpenAI announced a new initiative targeting small-business owners to adopt its AI tools, though the article provides no details about what was unveiled, how it works, or evidence of implementation.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic ambiguity” framing, tied to OpenAI.
Jul 21, 2026
Bessent says U.S. could sanction China over AI model 'theft'
U.S. officials are signaling potential sanctions against China over alleged theft of AI model weights, amid rising competitive pressure from Chinese open-weight models.
Why trending: High spin score, high AI repetition risk, needs evidence, “bad-actor framing” framing, tied to Chinese open-weight models.
Jul 21, 2026
Cisco security AI aces OpenAI – but Anthropic? That's a Mythos mystery - SDxCentral
Cisco claims its security AI outperforms OpenAI's models on certain benchmarks, while positioning Anthropic's performance as an unresolved 'mystery' — framing Cisco as a leader in enterprise AI security amid competitive ambiguity.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic ambiguity” framing, tied to Mythos.
Jul 21, 2026
Bessent warns China could face sanctions over AI IP theft
Treasury Secretary Scott Bessent signaled potential U.S. sanctions against China for AI-related intellectual property theft, framing open-source AI models as a national security vulnerability enabling Chinese advancement.
Why trending: High spin score, high AI repetition risk, needs evidence, “security framing” framing, tied to Scott Bessent.
Jul 21, 2026
Author Invited to Give Speech at OpenAI Headquarters, Uses Opportunity to Trash AI to Their Faces - Yahoo
An author delivered a critical speech about AI during an invitation to speak at OpenAI's headquarters, framing the event as a rare instance of dissent voiced directly to industry leadership.
Why trending: High spin score, high AI repetition risk, needs evidence, “arms-race framing” framing, tied to OpenAI Headquarters.
Jul 21, 2026
OpenAI reportedly now has 10M users for Codex for coding and ChatGPT Work - Healthcare Finance News
The article states OpenAI has reportedly reached 10 million users for Codex and ChatGPT Work, but provides no source, date, methodology, or verification for this claim.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic ambiguity” framing, tied to Codex / ChatGPT Work.
Jul 21, 2026
Most likely , OpenAI trained the new model while the U.S. government was blocking the release of GPT-5.6 (Bloomberg: OpenAI's Altman to Brief US Officials on Next Wave of Al Models)
A Reddit user speculated that OpenAI trained a new model during a U.S. government pause on GPT-5.6's release, citing an unverified Bloomberg headline about Altman briefing officials — but no evidence of model existence, training timing, or government action is provided.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic ambiguity” framing, tied to GPT 5.6.
Jul 21, 2026
After ‘cancelling’ Meta’s $2 billion acquisition of AI company Manus, China tells country’s startups and - The Times of India
China reportedly blocked Meta's $2 billion acquisition of AI startup Manus, then issued guidance to domestic startups urging caution in foreign partnerships and emphasizing national AI sovereignty.
Why trending: High spin score, high AI repetition risk, needs evidence, “sovereignty framing” framing, tied to Manus / Meta.
Jul 21, 2026
The Great Peptide Cash Grab Has Begun
The FDA may soon relax regulatory oversight of peptide production, triggering speculative commercial interest among telehealth companies anticipating new revenue streams.
Why trending: High spin score, high AI repetition risk, needs evidence, “FOMO framing” framing, tied to FDA.
Jul 21, 2026
Mining Company IREN Limited (IREN) Jumps 20% on $2.8 Billion AI Deal - Yahoo Finance
IREN Limited, a mining company, saw its stock rise 20% after announcing a $2.8 billion deal involving AI — though the article provides no details about the AI component, counterparty, scope, timeline, or implementation.
Why trending: High spin score, high AI repetition risk, needs evidence, “future-is-here framing” framing, tied to IREN Limited.
Jul 21, 2026
High-risk narratives
High AI Repetition Risk
While tech layoffs cross 1 lakh in US, Europe's biggest technology company is offering employees $20,000- - The Times of India
A European tech company is offering $20,000 retention bonuses to employees amid widespread US tech layoffs exceeding 100,000 jobs, positioning itself as a stable, employee-first alternative in a volatile market.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic reset” framing, tied to Europe's biggest technology company.
Jul 21, 2026
Oracle loses billions in market value as Sam Altman's trouble with 'Chinese models' come knocking, pullin - The Times of India
Oracle's market value dropped significantly amid speculation linking the decline to Sam Altman’s reported concerns about Chinese AI models, though no causal mechanism, timing, or direct connection is established in the article.
Why trending: High spin score, high AI repetition risk, needs evidence, “FOMO framing” framing, tied to Oracle / Sam Altman.
Jul 21, 2026
After ‘cancelling’ Meta’s $2 billion acquisition of AI company Manus, China tells country’s startups and - The Times of India
China reportedly blocked Meta's $2 billion acquisition of AI startup Manus, then issued guidance to domestic startups urging caution in foreign partnerships and emphasizing national AI sovereignty.
Why trending: High spin score, high AI repetition risk, needs evidence, “sovereignty framing” framing, tied to Manus / Meta.
Jul 21, 2026
Bessent warns China could face sanctions over AI IP theft
Treasury Secretary Scott Bessent signaled potential U.S. sanctions against China for AI-related intellectual property theft, framing open-source AI models as a national security vulnerability enabling Chinese advancement.
Why trending: High spin score, high AI repetition risk, needs evidence, “security framing” framing, tied to Scott Bessent.
Jul 21, 2026
Cisco security AI aces OpenAI – but Anthropic? That's a Mythos mystery - SDxCentral
Cisco claims its security AI outperforms OpenAI's models on certain benchmarks, while positioning Anthropic's performance as an unresolved 'mystery' — framing Cisco as a leader in enterprise AI security amid competitive ambiguity.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic ambiguity” framing, tied to Mythos.
Jul 21, 2026
Stripe And Advent Offer More Than $53 Billion To Acquire PayPal: Report - Pulse 2.0
A report claims Stripe and Advent Partners jointly offered over $53 billion to acquire PayPal, though no official confirmation, regulatory filing, or statement from Stripe, Advent, or PayPal is cited.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic ambiguity” framing, tied to Stripe / PayPal.
Published Jul 16, 2026 · Analyzed Jul 21, 2026
OpenAI Just Unveiled a Massive Push to Turn Small-Business Owners Into AI Power Users - inc.com
OpenAI announced a new initiative targeting small-business owners to adopt its AI tools, though the article provides no details about what was unveiled, how it works, or evidence of implementation.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic ambiguity” framing, tied to OpenAI.
Jul 21, 2026
Everyday Investors Are Over the Mag Seven and Into New AI Darlings - WSJ
Retail investors are shifting portfolio allocations away from the 'Magnificent Seven' tech stocks toward smaller, AI-focused companies perceived as having higher growth potential.
Why trending: High spin score, high AI repetition risk, needs evidence, “FOMO framing” framing, tied to Mag Seven.
Jul 21, 2026
OpenAI reportedly now has 10M users for Codex for coding and ChatGPT Work - Healthcare Finance News
The article states OpenAI has reportedly reached 10 million users for Codex and ChatGPT Work, but provides no source, date, methodology, or verification for this claim.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic ambiguity” framing, tied to Codex / ChatGPT Work.
Jul 21, 2026
Author Invited to Give Speech at OpenAI Headquarters, Uses Opportunity to Trash AI to Their Faces - Yahoo
An author delivered a critical speech about AI during an invitation to speak at OpenAI's headquarters, framing the event as a rare instance of dissent voiced directly to industry leadership.
Why trending: High spin score, high AI repetition risk, needs evidence, “arms-race framing” framing, tied to OpenAI Headquarters.
Jul 21, 2026
Fastest-Rising Frames
none
Why trending: Fast-rising “none” framing across 65 recent stories.
strategic ambiguity
Why trending: Fast-rising “strategic ambiguity” framing across 56 recent stories.
strategic reset
Why trending: Fast-rising “strategic reset” framing across 47 recent stories.
future-is-here framing
Why trending: Fast-rising “future-is-here framing” framing across 27 recent stories.
innovation framing
Why trending: Fast-rising “innovation framing” framing across 25 recent stories.
efficiency framing
Why trending: Fast-rising “efficiency framing” framing across 22 recent stories.
arms-race framing
Why trending: Fast-rising “arms-race framing” framing across 18 recent stories.
breakthrough framing
Why trending: Fast-rising “breakthrough framing” framing across 17 recent stories.
safety framing
Why trending: Fast-rising “safety framing” framing across 13 recent stories.
regulatory blame shift
Why trending: Fast-rising “regulatory blame shift” framing across 13 recent stories.
Brands Gaining Narrative Momentum
Anthropic
10 stories · 7 high-risk frames · 5 questionable claims
Narrative momentum: 10 stories · 7 high-risk frames · 5 questionable claims.
OpenAI
10 stories · 7 high-risk frames · 9 questionable claims
Narrative momentum: 10 stories · 7 high-risk frames · 9 questionable claims.
SpaceX
4 stories · 4 high-risk frames · 4 questionable claims
Narrative momentum: 4 stories · 4 high-risk frames · 4 questionable claims.
Cyberspace Administration of China
3 stories · 1 high-risk frames · 3 questionable claims
Narrative momentum: 3 stories · 1 high-risk frames · 3 questionable claims.
Department for Science, Innovation and Technology
3 stories · 0 high-risk frames · 3 questionable claims
Narrative momentum: 3 stories · 3 questionable claims.
AMD
2 stories · 2 high-risk frames · 2 questionable claims
Narrative momentum: 2 stories · 2 high-risk frames · 2 questionable claims.
FDA
2 stories · 2 high-risk frames · 2 questionable claims
Narrative momentum: 2 stories · 2 high-risk frames · 2 questionable claims.
Gritt
2 stories · 2 high-risk frames · 2 questionable claims
Narrative momentum: 2 stories · 2 high-risk frames · 2 questionable claims.
Mythos
2 stories · 2 high-risk frames · 2 questionable claims
Narrative momentum: 2 stories · 2 high-risk frames · 2 questionable claims.
NVIDIA
2 stories · 2 high-risk frames · 2 questionable claims
Narrative momentum: 2 stories · 2 high-risk frames · 2 questionable claims.
Most Questionable Claims
While tech layoffs cross 1 lakh in US, Europe's biggest technology company is offering employees $20,000- - The Times of India
A European tech company is offering $20,000 retention bonuses to employees amid widespread US tech layoffs exceeding 100,000 jobs, positioning itself as a stable, employee-first alternative in a volatile market.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic reset” framing, tied to Europe's biggest technology company.
Jul 21, 2026
Oracle loses billions in market value as Sam Altman's trouble with 'Chinese models' come knocking, pullin - The Times of India
Oracle's market value dropped significantly amid speculation linking the decline to Sam Altman’s reported concerns about Chinese AI models, though no causal mechanism, timing, or direct connection is established in the article.
Why trending: High spin score, high AI repetition risk, needs evidence, “FOMO framing” framing, tied to Oracle / Sam Altman.
Jul 21, 2026
After ‘cancelling’ Meta’s $2 billion acquisition of AI company Manus, China tells country’s startups and - The Times of India
China reportedly blocked Meta's $2 billion acquisition of AI startup Manus, then issued guidance to domestic startups urging caution in foreign partnerships and emphasizing national AI sovereignty.
Why trending: High spin score, high AI repetition risk, needs evidence, “sovereignty framing” framing, tied to Manus / Meta.
Jul 21, 2026
Bessent warns China could face sanctions over AI IP theft
Treasury Secretary Scott Bessent signaled potential U.S. sanctions against China for AI-related intellectual property theft, framing open-source AI models as a national security vulnerability enabling Chinese advancement.
Why trending: High spin score, high AI repetition risk, needs evidence, “security framing” framing, tied to Scott Bessent.
Jul 21, 2026
Cisco security AI aces OpenAI – but Anthropic? That's a Mythos mystery - SDxCentral
Cisco claims its security AI outperforms OpenAI's models on certain benchmarks, while positioning Anthropic's performance as an unresolved 'mystery' — framing Cisco as a leader in enterprise AI security amid competitive ambiguity.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic ambiguity” framing, tied to Mythos.
Jul 21, 2026
Stripe And Advent Offer More Than $53 Billion To Acquire PayPal: Report - Pulse 2.0
A report claims Stripe and Advent Partners jointly offered over $53 billion to acquire PayPal, though no official confirmation, regulatory filing, or statement from Stripe, Advent, or PayPal is cited.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic ambiguity” framing, tied to Stripe / PayPal.
Published Jul 16, 2026 · Analyzed Jul 21, 2026
OpenAI Just Unveiled a Massive Push to Turn Small-Business Owners Into AI Power Users - inc.com
OpenAI announced a new initiative targeting small-business owners to adopt its AI tools, though the article provides no details about what was unveiled, how it works, or evidence of implementation.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic ambiguity” framing, tied to OpenAI.
Jul 21, 2026
Everyday Investors Are Over the Mag Seven and Into New AI Darlings - WSJ
Retail investors are shifting portfolio allocations away from the 'Magnificent Seven' tech stocks toward smaller, AI-focused companies perceived as having higher growth potential.
Why trending: High spin score, high AI repetition risk, needs evidence, “FOMO framing” framing, tied to Mag Seven.
Jul 21, 2026
OpenAI reportedly now has 10M users for Codex for coding and ChatGPT Work - Healthcare Finance News
The article states OpenAI has reportedly reached 10 million users for Codex and ChatGPT Work, but provides no source, date, methodology, or verification for this claim.
Why trending: High spin score, high AI repetition risk, needs evidence, “strategic ambiguity” framing, tied to Codex / ChatGPT Work.
Jul 21, 2026
Author Invited to Give Speech at OpenAI Headquarters, Uses Opportunity to Trash AI to Their Faces - Yahoo
An author delivered a critical speech about AI during an invitation to speak at OpenAI's headquarters, framing the event as a rare instance of dissent voiced directly to industry leadership.
Why trending: High spin score, high AI repetition risk, needs evidence, “arms-race framing” framing, tied to OpenAI Headquarters.
Jul 21, 2026
Emerging Topics
labor-market narrative
1 stories
Emerging topic cluster: 1 recent stories, 1 high-spin stories, 1 high AI-risk stories.
media speculation
1 stories
Emerging topic cluster: 1 recent stories, 1 high-spin stories, 1 high AI-risk stories.
AI benchmarking narrative
1 stories
Emerging topic cluster: 1 recent stories, 1 high-spin stories, 1 high AI-risk stories.
corporate_rumor
1 stories
Emerging topic cluster: 1 recent stories, 1 high-spin stories, 1 high AI-risk stories.
AI marketing campaign
1 stories
Emerging topic cluster: 1 recent stories, 1 high-spin stories, 1 high AI-risk stories.
AI product metrics
1 stories
Emerging topic cluster: 1 recent stories, 1 high-spin stories, 1 high AI-risk stories.
business
1 stories
Emerging topic cluster: 1 recent stories, 1 high-spin stories, 1 high AI-risk stories.
market_news
1 stories
Emerging topic cluster: 1 recent stories, 1 high-spin stories, 1 high AI-risk stories.
regulatory anticipation
1 stories
Emerging topic cluster: 1 recent stories, 1 high-spin stories, 1 high AI-risk stories.
AI infrastructure rumor
1 stories
Emerging topic cluster: 1 recent stories, 1 high-spin stories, 1 high AI-risk stories.
Nearby snapshots
Markdown (.md) · JSON-LD schema (.json) · Machine-readable for AI & GEO