AI News Today July 8 2026: 15 Biggest Stories

July 8, 2026 is the day Fable 5 becomes a paid add-on for everyone. It is also the day CNBC confirmed that 30 to 46% of enterprise AI token usage at US companies is flowing to Chinese models. And it is the day Trump cancelled an AI executive order signing ceremony without explanation. Here are the 15 stories that define today. For continuous daily coverage of the full AI frontier, the AI Industry News and Trends hub at Build Fast with AI is your running reference.

1. Fable 5 Usage Credits Begin Today: The Full Billing Breakdown for Every Subscription Tier

Starting today, July 8, 2026, every Claude user who accesses Fable 5 is paying usage credits. Here is the full billing breakdown by tier. Claude Pro ($17/month): Fable 5 is available through credit purchases at $10 per million input tokens and $50 per million output tokens. Sonnet 5 ($2/$10 introductory through August 31) and Opus 4.8 ($5/$25) remain included in the Pro subscription. Claude Max ($100/month): same credit structure for Fable 5. Claude Max provides higher usage limits on included models. Team and Business plans: same credit structure. Enterprise seat plans: Fable 5 has never been included in seat pricing; all usage has always been credit-billed. The 50% weekly usage inclusion that Anthropic offered as a compensatory window during and after the 19-day export control suspension expired yesterday, July 7. What this means in practice: a single Fable 5 agentic coding session processing 2 million output tokens costs $100 in credits. The same session on Sonnet 5 at introductory pricing costs $20. The same session on Opus 4.8 costs $50. The performance gap between Fable 5 and Sonnet 5 on agentic coding is approximately 6 percentage points on SWE-bench Pro (63.2% Sonnet 5 vs 69.2% Opus 4.8 vs approximately 80%+ for Fable 5). For most enterprise coding workflows, Sonnet 5 at $2/$10 is the economically rational default; Fable 5 at $10/$50 is reserved for the hardest tasks where that performance gap is decisive. Audit your claude-fable-5 routing today and set credit limits before production traffic generates unexpected bills. Full billing documentation is at Anthropic's pricing page.

2. CNBC Confirms: Chinese AI Models Now 30-46% of US Enterprise Token Usage

CNBC published a major investigation on July 7, 2026 confirming that Chinese AI models now account for between 30% and 46% of the enterprise API token usage flowing through US developer platforms. The key data points: through OpenRouter, Chinese model share has been above 30% of all gateway tokens every week since February 8, 2026, rising as high as 46%. The average across the prior 12 months was just 11%, and had been as low as 4.5% in the first half of 2025. Through Vercel, DeepSeek saw its share of gateway tokens climb in the May-June period. Z.ai's GLM-5.2 saw the fastest adoption of any model tracked by Vercel in 2026: daily token volume grew approximately 27x and customer count grew approximately 80x in its first full week after launch. Justin Summerville at OpenRouter quantified the price advantage: open-source Chinese models are 60 to 90 percent cheaper than leading Anthropic and OpenAI models. Harpreet Arora at Vercel described the driver bluntly: 'Price is doing the work here. When a task doesn't need the best model, teams are beginning to route it to the cheapest one that's good enough, and the recent wave of models coming out of China is winning that trade.' GLM-5.2 landed within one percentage point of Anthropic's Opus 4.8 on one closely watched agentic benchmark, at roughly a fifth of the cost. The structural implication: US frontier AI labs are pricing themselves out of the middle tier of enterprise workflows, ceding that market to Chinese open-weight models, while defending the top-tier performance market where cost is secondary to capability. The CNBC investigation on Chinese AI model adoption has the full data breakdown.

3. Why US Engineers Are Choosing GLM-5.2 and DeepSeek Over Claude and GPT

The shift to Chinese AI models is not primarily ideological or geopolitical. It is economic and technical. The engineer's calculation is simple: for tasks that do not require frontier-class performance (routine summarization, code completion, data extraction, customer support drafting), a model that costs 60 to 90 percent less and performs within 5 to 10 percentage points of the frontier is the rational choice. The advisor model technique that has been gaining adoption since Q2 2026, where a cheap open-weight model handles the bulk of tasks and escalates to a frontier model only when needed, makes Chinese models the natural default tier for most enterprise AI routing. GLM-5.2 specifically has acquired a reputation for strong agentic coding performance: it scored 62.1% on SWE-bench Pro (above GPT-5.5 at 58.6%) and carries a MIT license. Z.ai's MIT licensing language specifically says 'no regional limits,' which is particularly relevant in the post-Fable-5-export-control environment, where developers are actively evaluating which models they can rely on regardless of US geopolitical decisions. Yacine Jernite at Hugging Face: 'We are seeing companies increasingly motivated to turn to cheaper AI stacks they can control and adapt themselves, and given the state of open-source and open-weight models that often means leveraging Chinese options.' The counter-consideration: data jurisdiction (API calls route through Chinese servers), content restrictions on politically sensitive topics, and tool-call schema reliability gaps. For enterprise teams handling regulated data or sensitive business intelligence, the data jurisdiction issue is not minor.

4. Trump Cancels AI Executive Order Signing Ceremony: What Happened and What Comes Next

President Trump abruptly cancelled a scheduled Oval Office signing ceremony for a new AI executive order, according to reporting from the AI news roundup aggregated by crescendo.ai and multiple news sources. Trump told reporters he did not want to do anything that would interfere with the US competitive position in AI, citing concern that signing would 'undermine America's lead over China.' The proposed order had emerged from growing pressure within the banking and financial sectors over AI cybersecurity risks, particularly those raised by Claude Mythos model capabilities. The cancellation came during the same window the White House was expected to announce the voluntary AI model standards framework (July 7-11, per FT reporting from July 2). The relationship between the cancelled executive order and the voluntary standards framework is unclear: they may be separate instruments that were conflated in pre-announcement reporting, or the cancellation of the signing ceremony may reflect internal White House disagreement about whether any AI regulatory action risks the competitive advantage narrative. The practical consequence: the August 1 formal deadline for NSA and CISA to deliver the classified frontier model benchmarks and the voluntary pre-release framework still stands regardless of whether a new executive order is signed. That deadline is not contingent on a signing ceremony.

5. The August 1 Framework Deadline Is Now the Only Firm Governance Anchor Left

With the White House signing ceremony cancelled and the voluntary standards announcement window having passed without a confirmed public announcement, the August 1, 2026 deadline established in Trump's June 2 executive order is now the only confirmed, immovable governance milestone in the frontier AI release calendar. By August 1, the NSA (in consultation with ONCD, CISA, and the Department of War) must deliver a classified benchmarking process to assess the advanced cyber capabilities of frontier AI models and determine which models qualify as 'covered frontier models' triggering the voluntary pre-release review window. The same agencies must make the voluntary engagement framework available to frontier model developers. These are statutory deliverables under the executive order, not aspirational targets that can slip based on political calendar. Whether the White House announces the framework publicly before August 1 (creating an opportunity for OpenAI, Anthropic, and Google to align their July model launches with the framework) or simply delivers it on August 1 with less fanfare than anticipated determines the practical rollout path for GPT-5.6 broad access and Gemini 3.5 Pro. For the full executive order text and implementation timeline, the White House EO on Promoting AI Innovation and Security is the primary reference.

6. GPT-5.6 Broad Release Path: What the White House Uncertainty Means for ChatGPT and API Access

GPT-5.6 Sol, Terra, and Luna remain limited to approximately 20 government-vetted partner organizations as of July 8. The White House uncertainty created by the cancelled signing ceremony affects the broad release path in one specific way: OpenAI needs the voluntary standards framework to be formally established before it can publicly frame GPT-5.6's broad release as compliant with the government's own stated process. Without that framework published, OpenAI expanding access would look like unilateral action on a model the government has asked to be kept gated, which carries export control risk analogous to what Anthropic faced with Fable 5. The most likely resolution: OpenAI and the White House negotiate a quiet expansion of the vetted partner list over the next week, adding enough enterprise customers to demonstrate momentum without requiring a formal policy announcement. The August 1 deadline then formalizes the framework retroactively. Former White House AI adviser Dean Ball, now joining OpenAI, argued that the current de facto involuntary licensing regime for frontier AI risks handing China an advantage. If GPT-5.6 remains gated past mid-July, his argument becomes the dominant narrative in the AI policy debate.

7. Alberta Government Uses Claude for Cybersecurity: The First Non-US Government AI Security Case Study

Anthropic published on July 6, 2026 a case study documenting the Government of Alberta's use of Claude to find and fix cybersecurity vulnerabilities across government systems. Alberta becomes the first Canadian provincial government to publish a formal AI cybersecurity case study, extending Anthropic's government security program beyond its existing US partnerships. The Alberta deployment integrates Claude with a security scanning workflow that identifies vulnerabilities in government code repositories, prioritizes findings by severity and exploitability, generates candidate patches for high-priority issues, and produces audit-ready remediation documentation for the security team. Alberta reported significantly reduced mean time to remediation for identified vulnerabilities compared to its previous manual security review process, with the AI-assisted approach enabling security staff to review and validate more findings in the same time period. The case study comes one week after the JADEPUFFER AI ransomware disclosure (July 6) and in the same news cycle as the Five Eyes warning (June 22) that AI cyberattacks on governments are months away. Anthropic is positioning the Alberta deployment as proof of its defensive AI security thesis: the same frontier model capabilities that make offensive AI cybersecurity threats real are also the best tool for defending against them. For the full context on Anthropic's government security programs, the AI industry news hub at Build Fast with AI covers Project Glasswing, Patch the Planet, and the full government AI security landscape.

8. Thrive Holdings Raises $2 Billion to Acquire and Transform Professional Services Firms With AI

The Information reported that Thrive Holdings, a one-year-old holding company started by OpenAI investor Thrive Capital, is raising approximately $2 billion from investors including Altimeter Capital, D1 Capital Partners, and SoftBank. Thrive Holdings' strategy: buy controlling stakes in accounting, legal, and other professional services firms and transform them with AI. This is a direct implementation bet on the AI disruption of professional services, separate from investing in AI labs. The thesis is straightforward: accounting and legal firms are the highest-value knowledge work category most directly threatened by frontier AI, but their transformation timeline is constrained by regulatory requirements, partner governance structures, and client trust requirements that make organic AI adoption slow. A holding company with controlling stakes can implement AI transformation across multiple firms simultaneously, capturing the productivity gains (and margin expansion) that AI enables in professional services workflows. The $2 billion raise would give Thrive Holdings the capital to acquire controlling stakes in firms with combined annual revenues in the tens of billions. Altimeter, D1, and SoftBank's participation signals institutional conviction that the professional services AI disruption is a fundable investment thesis, not just a theoretical concern.

9. Qualcomm-Tenstorrent Acquisition Talks: Jim Keller Denied It, What the RISC-V Story Actually Is

The Qualcomm-Tenstorrent acquisition story has a clear resolution as of July 8. The Information reported on June 15 that Qualcomm was in talks to acquire Tenstorrent at $8-10 billion. On June 30, Tenstorrent CEO Jim Keller publicly denied the talks, stating the company has not been in acquisition discussions and will focus on its own business. Qualcomm shares fell approximately 1% after the original report and have since normalized. The underlying RISC-V chip story remains significant regardless of the acquisition denial. Tenstorrent's Galaxy Blackhole platform, launched in 2026, features 32 accelerators with 768 RISC-V cores each in a 6U data center enclosure. RISC-V is an open, royalty-free chip architecture that allows any company to build processors without paying ARM or x86 licensing fees, which is increasingly attractive to hyperscalers trying to control their own compute destiny. Qualcomm has already moved in this direction through its December 2025 acquisition of Ventana Micro Systems (another RISC-V designer) and its $3.92 billion acquisition of Modular (the AI model deployment platform). Keller's denial ends the specific acquisition story, but Qualcomm's RISC-V push and Tenstorrent's independent trajectory as a data center AI chip company both continue. For the full AI chip landscape, the AI industry news hub at Build Fast with AI covers Nvidia, Broadcom, Amazon Trainium, OpenAI Jalapeño, and the full AI chip stack.

10. Anthropic Updates Privacy Policy for Government ID Verification: What It Means for Fable 5 Access

Today, July 8, 2026, Anthropic's updated privacy policy requiring government-issued ID verification takes effect, per the timeline Anthropic established as part of the Fable 5 export control redeployment agreement. The ID verification requirement is the mechanism through which Anthropic is implementing the US government's condition that Fable 5 access be confirmed as not reaching foreign nationals in violation of the export control terms that were in force from June 12 to June 30. In practice: users accessing Fable 5 through claude.ai subscriptions may be asked to verify their identity with a government-issued ID if their account profile does not already meet the verification criteria. Enterprise users accessing through the API who have not already completed enterprise KYC verification may face similar prompts. The verification requirement is most likely to affect users whose account indicators (time zone, payment method, IP address patterns) suggest a location that might trigger the original export control concern. For the vast majority of US-based subscribers, the ID verification is a one-time check that does not affect ongoing usage. For international subscribers, the policy change signals that Anthropic is implementing systematic verification infrastructure, not just relying on terms-of-service agreements.

11. GLM-5.2 Adoption Surge: 80x Customer Growth on Vercel in First Week, 27x Daily Token Volume

The CNBC July 7 investigation contains specific Vercel platform data on GLM-5.2's adoption that quantifies exactly how fast a strong Chinese open-weight model gains enterprise adoption when pricing is competitive. In GLM-5.2's first full week after launch on Vercel: daily token volume grew approximately 27 times. The number of customers using it grew approximately 80 times. Vercel Head of Agentic Infrastructure Harpreet Arora confirmed these numbers to CNBC: 'In its first full week after launch, daily token volume grew about 27x and the number of customers using it grew about 80x.' For context: 80x customer growth in a single week is unprecedented for any model on the Vercel platform. The reason is the same every time: GLM-5.2 scored 62.1% on SWE-bench Pro (comparable to frontier Western models) at $1.40/$4.40 per million tokens (a fraction of Anthropic and OpenAI pricing). For developers using Vercel to deploy AI applications, the economics of GLM-5.2 relative to Claude Sonnet 5 ($2/$10) or GPT-5.5 ($2.50/$15) are compelling enough to drive immediate adoption regardless of the model's Chinese origin. The 80x customer growth statistic is the clearest quantified evidence yet of what the tokenmaxxing correction plus competitive open-weight Chinese models is doing to Western frontier lab market share among cost-sensitive enterprise developers. For the full model comparison including GLM-5.2, the best AI models July 2026 guide at Build Fast with AI has verified benchmarks and pricing.

12. The Tokenmaxxing Sequel: Enterprise AI Spending Patterns After the Q2 Budget Crisis

CNBC's July 7 investigation on Chinese model adoption is the latest chapter in the tokenmaxxing sequel story that began in Q2 2026. The sequence: Q1 2026, enterprise AI spending surged as the 'vibe coding' wave pushed developers to use frontier models for every task without cost controls. Uber burned through its entire 2026 annual AI budget in four months. Lindy CEO switched 100% off Claude to DeepSeek after costs became unsustainable. Q2 2026, the enterprise correction arrived: companies implemented per-employee spend tiers, usage analytics, and model routing controls. Claude Enterprise launched spend alerts and model-level entitlements. GitHub Copilot moved to usage-based billing, causing 10x-50x cost increases for power users. Q3 2026 opening, the correction is now producing structural market shifts: Chinese open-weight models capturing 30-46% of US enterprise token volume. GLM-5.2 achieving 80x customer growth in a single week. The advisor model technique becoming mainstream: cheap Chinese open-weight model as default, frontier Western model as exception. The frontier AI pricing dynamic: Western labs raised prices in 2026 (GPT-5.5 doubled GPT-5.4, Fable 5 at $10/$50, Gemini 3.5 Flash at 3x its predecessor) at exactly the moment Chinese open-weight models reached near-frontier performance at deeply discounted prices. The market correction is rational and now quantified.

13. Gemini 3.5 Pro: Still No Launch Date, Still the Most Anticipated Model of July

As of July 8, 2026, Gemini 3.5 Pro has not achieved general availability. The model remains in limited Vertex AI enterprise preview. There is no confirmed launch date. The most anticipated frontier model launch of July 2026 is now in its third week of delay past the June 30 target that was itself a month-late slip from the I/O commitment. Google has said it needs additional time to address token efficiency issues, coding performance gaps, and long-task reasoning gaps flagged by early enterprise testers. The competitive context is challenging: GPT-5.6 is gated but confirmed to be launching broadly soon. Claude Sonnet 5 launched June 30 with near-Opus performance at Sonnet pricing. Gemini 2.5 Pro with Deep Think launched June 22 with the strongest science and reasoning benchmark scores ever published. Gemini 3.5 Pro needs to deliver on its confirmed specifications (2-million-token context window, Deep Think reasoning, frontier multimodal) while also demonstrating that Google has closed the coding performance gap with Anthropic and OpenAI that its own Flash model highlighted. Every day of additional delay strengthens the narrative that Google's talent departures have affected its execution cadence. The July Gemini 3.5 Pro launch, when it comes, will be one of the most scrutinized model launches of 2026.

14. Bespoke Labs Raises $40 Million for AI Post-Training Infrastructure

AI post-training startup Bespoke Labs raised $40 million in a new funding round, per SiliconANGLE reporting on July 7, 2026. Bespoke Labs builds infrastructure for the post-training phase of AI model development, specifically the reinforcement learning from human feedback (RLHF), preference data collection, and fine-tuning pipeline layers that determine how a model behaves after initial pretraining. The post-training market is strategically important because it is where the commercial differentiation between frontier models happens: two models with similar pretraining can produce dramatically different behaviors based on post-training quality. Bespoke Labs targets both frontier labs doing model development and enterprises doing custom fine-tuning of foundation models for specific domains. The $40 million raise at this scale implies a valuation in the $150-200 million range, modest by frontier AI lab standards but consistent with infrastructure tooling companies serving the ecosystem. The post-training infrastructure market is growing as the number of organizations doing domain-specific fine-tuning on open-weight models (including GLM-5.2, LongCat-2.0, and the Llama family) expands. For the developer tools and infrastructure ecosystem context, the AI coding tools hub at Build Fast with AI covers the full developer tooling landscape from model training to deployment.

15. The New Frontier AI Governance Calendar: Every Key Date Through August 2026

July 8 through August 1 is the most consequential governance window in frontier AI history. Here is the complete calendar of key dates through August 1, 2026. July 8 (today): Fable 5 usage credits billing begins; Anthropic ID verification privacy policy takes effect. July 8-11: White House voluntary AI standards framework announcement window (status uncertain after signing ceremony cancellation). July 15: China AI companion law enforcement deadline; Doubao and Qwen agent features shut down; Doubao user agent data export window closes. July 15: Claude Science AI for Science grants application deadline closes. July 31: Claude AI Science grant award notifications sent. August 1: Formal NSA and CISA deadline to deliver classified frontier model benchmarking process and voluntary pre-release framework under Trump's June 2 executive order. August 1 deadline triggers: the formal framework for determining which models qualify as covered frontier models, the repeatable pre-release review process for future model releases, the international access rules for frontier AI models. August 31: Claude Sonnet 5 introductory pricing ($2/$10 per million tokens) expires; standard pricing ($3/$15) takes effect. By September: OpenAI targeting IPO roadshow. By October: Anthropic targeting IPO roadshow. Q3 2026: Grok 4.5 anticipated public release; Grok 5 continues training on Colossus 2. Gemini 3.5 Pro general availability (no confirmed date). GPT-5.6 broad ChatGPT and API access (dependent on government framework). These dates collectively define the governance, commercial, and product landscape for the next six weeks of the AI industry.

Frequently Asked Questions

How much does Fable 5 cost for a Pro subscriber starting July 8?

A Claude Pro subscriber ($17/month) who uses Fable 5 starting July 8 will be billed usage credits at $10 per million input tokens and $50 per million output tokens on top of the Pro subscription. A single agentic coding session processing 2 million output tokens costs $100 in credits. Sonnet 5 at introductory pricing ($2/$10 through August 31) costs $20 for the same session. Opus 4.8 ($5/$25) costs $50. Both Sonnet 5 and Opus 4.8 remain included in Pro plan usage limits without additional credit charges.

Is using Chinese AI models through OpenRouter safe for enterprise data?

Using Chinese models through OpenRouter provides some protection because OpenRouter acts as an intermediary, but the ultimate data path and terms of service for each underlying model still apply. For regulated data (healthcare, financial, legal), the safest path for Chinese models is through Azure AI (which hosts DeepSeek and Qwen models under Microsoft's data processing agreements) or Cloudflare Workers AI. Direct API calls to Z.ai, DeepSeek, or Moonshot AI route through Chinese-jurisdiction servers, which typically violates data residency requirements in regulated industries.

What is the Bespoke Labs funding for?

Bespoke Labs raised $40 million to expand its post-training infrastructure platform, which builds RLHF pipelines, preference data collection tools, and fine-tuning infrastructure for both frontier AI labs and enterprises doing domain-specific model customization. The company is targeting the growing market of organizations that run custom fine-tuning on open-weight foundation models including GLM-5.2, LongCat-2.0, and Llama-family models for specific industry applications.

What is the Alberta government AI security deployment using Claude?

The Alberta government uses Claude through a security scanning workflow that scans government code repositories for vulnerabilities, prioritizes findings by severity and exploitability, generates candidate patches for high-priority issues, and produces audit-ready remediation documentation. Alberta reported significantly reduced mean time to remediation compared to its previous manual security review process. The deployment is part of Anthropic's broader government cybersecurity program that includes Project Glasswing in the United States and the Patch the Planet initiative with Trail of Bits and HackerOne.

What happened to the Qualcomm-Tenstorrent acquisition?

The Information reported on June 15 that Qualcomm was in talks to acquire Tenstorrent at $8-10 billion. Jim Keller, Tenstorrent's CEO, publicly denied the talks on June 30, saying the company has not been in acquisition discussions. The denial effectively ended the acquisition story. Qualcomm has pursued its AI chip strategy through other acquisitions: Modular ($3.92 billion, confirmed) and Ventana Micro Systems (another RISC-V designer, completed December 2025). Tenstorrent continues as an independent company focused on its Galaxy Blackhole RISC-V AI accelerator platform.

Why did Trump cancel the AI executive order signing ceremony?

Trump cancelled a scheduled Oval Office signing ceremony for a new AI executive order, telling reporters he did not want to do anything that would interfere with the US competitive position in AI. The proposed order had emerged from banking and financial sector pressure over AI cybersecurity risks. Trump's stated concern was that signing the order 'could undermine America's lead over China' in AI. The cancellation leaves the August 1 deadline under his June 2 executive order as the only confirmed governance milestone.

When will Gemini 3.5 Pro launch?

As of July 8, 2026, Gemini 3.5 Pro has no confirmed launch date. The model remains in limited Vertex AI enterprise preview. Google confirmed delays citing token efficiency issues and coding performance gaps flagged by enterprise testers. The July 2026 general availability window is now the expected range, but no specific date has been announced. Gemini 2.5 Pro with Deep Think, a separate model in the Gemini family, is currently available through the Gemini API and AI Studio with strong science and reasoning benchmarks.

Recommended Blogs

• AI News Today July 7 2026: JADEPUFFER First Autonomous Ransomware, Anthropic Overtakes OpenAI on Revenue
• AI News Today July 6 2026: Gemini 3.5 Pro in Preview, China AI Companion Law July 15, Tesla Robotaxi Miami
• AI News Today July 4 2026: Grok 4.5 Private Beta, LongCat-2.0, Pentagon Emails, Crunchbase VC Record
• Best AI Models July 2026: Full Ranked Leaderboard by Use Case
• Claude Code vs Codex vs Cursor: AI Coding Tools 2026 Full Comparison
• AI Industry News and Trends Hub: Running Daily Coverage of 2026

Resources and Community

Join our community of 70,000+ AI enthusiasts and learn to build powerful AI applications! Whether you are a beginner or an experienced developer, Build Fast with AI helps you understand and implement AI in your projects.

• Website — buildfastwithai.com
• LinkedIn — Build Fast with AI
• Instagram — @buildfastwithai
• Founder Twitter — @satvikps
• Twitter — @BuildFastWithAI

Agentic AI Launchpad 2026

A structured 6-week cohort program that takes you from AI basics to building and deploying real-world agentic AI systems. Includes live sessions, expert mentorship, project reviews, and a builder community network.

Ready to go from learning to building? Join the next cohort: Agentic AI Launchpad 2026

Free AI Resources

Access free tools, workshops, and micro-learning to keep building:

• AI Workshops — Free resources, upcoming events and past recordings
• Unrot — Learn AI in 5 minutes a day (free micro-learning app)

Subscribe to the Build Fast with AI newsletter and follow @BuildFastWithAI on X for sourced daily AI coverage across every frontier development.

References

• CNBC — Chinese AI Models Are Gaining Ground With US Companies as OpenAI and Anthropic Costs Surge
• Anthropic Newsroom — Government of Alberta Uses Claude to Find and Fix Cybersecurity Vulnerabilities
• Anthropic — More Details on Fable 5 Cyber Safeguards and Jailbreak Framework
• The Information via AI Weekly — Thrive Holdings Raises $2B to Transform Professional Services With AI
• AI Weekly — Qualcomm Targets Tenstorrent in Reported $10B RISC-V Deal
• Seeking Alpha — Tenstorrent CEO Jim Keller Denies Qualcomm Acquisition Talks
• Tom's Hardware — Qualcomm Mulls Taking Over Jim Keller's Tenstorrent at $8-10B
• GuruFocus — Qualcomm in Talks to Acquire AI Chip Startup Tenstorrent for $8-10 Billion
• Crescendo AI — Trump Abruptly Cancels AI Executive Order Signing Ceremony
• SiliconANGLE — Bespoke Labs Raises $40M in Funding for AI Post-Training Infrastructure
• White House — Executive Order Promoting AI Innovation and Security June 2 2026
• TechCrunch — OpenAI Limits GPT-5.6 Rollout After Government Request
• Fortune — Anthropic Overtakes OpenAI on Revenue, Sam Altman Seeks New World Order for AI
• Build Fast with AI — Best AI Models July 2026 Full Leaderboard

Build Fast with AI — AI News Today July 7 2026