Wix reported third-quarter 2026 revenue of $570.17 million and a net loss of $13.7 million, reaffirming its full-year revenue guidance and forecasting low double-digit growth in the fourth quarter. Its Base44 AI platform is nearing $250 million in annual recurring revenue, while new Studio Two supports React workflows and external AI tools. Wix expects a lower free-cash-flow margin as it expands.
Cybersecurity firm CrowdStrike introduced tools to secure generative AI and autonomous agents, including Agentic Adversary Emulation powered by SafeMind, Agent Insider Threat Detection and an AI Kill Switch. New integrations and partnerships include AWS, Anthropic, Dynatrace, AppGate and Dell. The additions bring AI security, identity, cloud and SIEM capabilities into the Falcon platform; a sovereign-cloud deployment in Saudi Arabia also targets markets with strict data rules. CrowdStrike still needs to convert the offerings into sustained subscription growth while managing investment and execution risks.
Two articles compare Alphabet and Broadcom as AI investments. They identify Alphabet as the lower-risk, risk-adjusted choice: its November 12, 2026 share price was $298.38, trailing P/E was 15.4, and estimated fair-value upside was 9.1%. Alphabet reported FY2025 revenue of $402.84 billion across advertising, cloud and AI, and held more cash than debt. Its shares fell 15.33% over a month after October 28 results showed EPS of $2.12 against $3.02 expected; analyst EPS estimates also declined 4.79% over 30 days. Broadcom offers more direct exposure to custom AI chips and networking, with 16.2% estimated upside but a higher 45.2 trailing P/E.
Japan-based Sakana AI is building a defense-sector push with hires Michael Rogers and Tom Moran. Rogers served 37 years in the U.S. Navy and led the NSA and U.S. Cyber Command from 2014 to 2018. Moran spent nearly 30 years at the CIA, including six assignments as Chief of Station across several regions; he is also leading an intelligence project at UC Berkeley.
Workplace examples point to both benefits and limits of AI. A study of 5,179 customer-support agents found generative AI raised issues resolved per hour by 14% on average and 34% for novice and less-skilled workers. One workplace-learning article urges teams to define desired behavior changes, identify barriers and set a result that would justify stopping an investment. BDO, by contrast, reports employee-led adoption: nearly 3,500 colleagues responded to a survey, 87% said they already used AI, and staff submitted more than 11,000 use ideas. Before formal training, over 600 employees had built more than 1,200 agents; training AI champions helped agent numbers rise about 80% in two weeks. One employee updated nearly 2,000 design files in about a day, a task previously estimated to take 20 people close to a month.
Several articles describe public, political and governance questions around AI. One reports that Nvidia CEO Jensen Huang resigned amid disputes over data-center construction and growing political opposition, arguing that the industry may struggle to regain public goodwill. A separate report says Hugging Face CEO Clément Delangue pledged to keep backing open-source AI after Huang’s departure. Nvidia’s planned acquisition of Hugging Face remains scheduled to close in the first half of 2027, while the U.S. Department of Justice has asked about its competitive effects. Nvidia has also announced a $6 billion deal with coding startup Poolside to support an open AI ecosystem in the U.S. At the University of Kentucky, research administrator David Durman received a religious exemption from workplace AI use, saying that personally reasoning through work and ethics is important to his moral development.
Other reports highlight reliability and security concerns. A Costco forecasting test used 214 filings, 3,000 timestamped facts and 16 GPT-OSS-120B runs costing 19 cents; five runs passed self-correction checks and predicted a $560 million two-year profit increase. The forecast appeared close, but two errors offset one another: the model included a membership-price increase after July 2024 while assuming executive memberships were 15% rather than the disclosed 45%. Correcting the latter changed the forecast error from 1.7% to 28%. Separately, researcher Jason Kao of Fog Security found a Google Gemini flaw that allowed downloads of images blocked by its filters, including potentially harmful or copyrighted material. The gap was between Gemini’s chat interface and Google’s file-delivery system, showing that safeguards can fail downstream of the model.
Key Takeaways
- Wix posted $570.17 million in third-quarter 2026 revenue and says its Base44 AI platform is nearing $250 million in annual recurring revenue.
- CrowdStrike launched AI-agent security tools, including an AI Kill Switch and SafeMind-powered adversary emulation, with integrations involving AWS, Anthropic, Dynatrace, AppGate and Dell.
- Alphabet is presented as the lower-risk AI stock choice, while Broadcom offers greater custom-chip and networking exposure at a higher valuation.
- Sakana AI hired former NSA and U.S. Cyber Command director Michael Rogers and former CIA officer Tom Moran for its defense push.
- A study of 5,179 support agents found generative AI increased issues resolved per hour by 14% overall and 34% for novice and less-skilled workers.
- BDO reports 87% of nearly 3,500 survey respondents already used AI; employee-built agents grew about 80% in two weeks after champion training.
- Hugging Face CEO Clément Delangue reaffirmed support for open-source AI; Nvidia’s planned acquisition of the company is slated to close in the first half of 2027, amid DOJ inquiries.
- A University of Kentucky employee received a religious exemption from workplace AI use, citing the importance of personal ethical judgment.
- A Costco model forecast’s apparent 1.7% error rose to 28% after correcting an incorrect executive-membership assumption that had offset another error.
- A Google Gemini flaw let users download images its filters blocked because of a gap between the chat interface and file-delivery system.
Wix deepens AI push as quarterly revenue reaches $570 million
Wix.com reported third quarter 2026 revenue of US$570.17 million and a net loss of US$13.7 million. The company reaffirmed its full year revenue guidance and expects low double digit revenue growth in the fourth quarter. It is investing in its Base44 AI platform, which is nearing US$250 million in annual recurring revenue, and rolling out Studio Two, which supports React workflows and external AI tools. Wix expects a lower free cash flow margin as it expands, while execution and spending remain key concerns.
CrowdStrike adds new tools to protect AI agents
CrowdStrike announced new products for securing generative AI and autonomous AI agents. Its launches include Agentic Adversary Emulation, powered by SafeMind, Agent Insider Threat Detection, and an AI Kill Switch. The company also announced AI security integrations with AWS, Anthropic, Dynatrace, and AppGate. The products and partnerships support CrowdStrike’s effort to expand its Falcon platform, though the company still needs to turn them into lasting revenue growth.
CrowdStrike expands Falcon with AI agent security tools and partnerships
CrowdStrike introduced AI focused security products and integrations for sovereign cloud, autonomous adversary testing, and AI agent controls. Its announcements include the AI Kill Switch and agent monitoring, along with partnerships involving AWS, Dell, Dynatrace, Anthropic, and others. The additions bring AI security, identity, cloud, and SIEM capabilities closer together in the Falcon platform and could support greater adoption of Falcon Flex. The company’s outlook depends on turning AI security into subscription growth while managing high expectations, investment costs, and execution risks. A CrowdStrike Falcon sovereign cloud deployment in Saudi Arabia also supports the company’s expansion into markets with strict data rules.
Alphabet offers lower risk AI exposure while Broadcom brings greater upside potential
The article names Alphabet Inc Class A as the best risk adjusted AI stock pick, citing its $298.38 share price on November 12, 2026, 15.4 times trailing P/E, and 9.1% estimated fair value upside. Alphabet reported FY2025 revenue of $402.84 billion, and its advertising, cloud, and AI products offer several ways to earn from AI. However, its shares fell 15.33% over one month after October 28 results showed EPS of $2.12, below the expected $3.02, and analyst EPS estimates fell 4.79% over 30 days. Broadcom is presented as a higher risk choice, with custom AI chips and networking, 16.2% fair value upside, and a 45.2 times trailing P/E. The article favors Alphabet for resilience and valuation, and Broadcom for greater exposure to AI infrastructure.
Alphabet leads on AI value while Broadcom offers a riskier growth path
The article identifies Alphabet Inc Class A as the best risk adjusted AI pick, with a share price of $298.38 on November 12, 2026, a trailing P/E of 15.4 times, and estimated fair value upside of 9.1%. Alphabet’s FY2025 revenue was $402.84 billion, supported by advertising, cloud, and AI products, and the company holds more cash than debt. Its shares fell 15.33% over one month after October 28 earnings showed EPS of $2.12 versus $3.02 expected, while analyst EPS estimates dropped 4.79% over 30 days. Broadcom is described as a higher risk alternative that benefits from custom AI chips and networking, with 16.2% fair value upside and a 45.2 times trailing P/E. The article favors Alphabet for resilience and valuation, and Broadcom for stronger AI infrastructure exposure with more risk.
Sakana AI brings former US intelligence chiefs into defense push
Sakana AI hired Michael Rogers and Tom Moran as it looks to sell its technology to the defense sector. Rogers spent 37 years in the U.S. Navy and led the NSA and U.S. Cyber Command from 2014 to 2018. Moran spent nearly 30 years at the CIA and served six times as Chief of Station in East Asia, Western Europe, the Middle East and Africa. The company says Rogers brings national security experience, while Moran is leading an intelligence project at UC Berkeley and stresses the need to make sound judgments quickly from large amounts of information.
Three questions can show whether workplace learning delivers results
AI can help workers finish tasks faster, but faster work does not always improve an organization. A study of 5,179 customer support agents found that generative AI increased the number of issues resolved per hour by 14 percent on average and by 34 percent for novice and less-skilled agents. The article says learning teams should ask three questions before starting an investment: What should people do differently, what is stopping them, and what result would make the organization stop? The author found that an AI training plan for client-facing consultants measured usage, confidence and drafting speed, but did not show whether proposals improved or set clear decisions for continuing the program.
Nvidia faces growing political backlash as AI troubles mount
The article says Nvidia CEO Jensen Huang resigned, adding to questions about the AI industry’s growing political and business challenges. It points to disputes over data center construction and reports that opponents of a Utah data center were attacked on Fox News by a millionaire. The article argues that Nvidia can replace its CEO, but the AI industry may struggle to regain public goodwill. It also says the fallout from AI has been building in Washington and state legislatures.
University of Kentucky worker wins religious exemption from workplace AI
David Durman, a research administrator at the University of Kentucky, received an exemption from using AI at work based on his religious beliefs. In an interview with This Morning, he said that thinking through his work, ethics and treatment of others helps him develop as a moral person. He argued that handing that work over to AI would weaken his ability to act morally. The article describes his non-theistic religious exemption as unusual and says his employer approved it.
Hugging Face CEO pledges stronger open source AI effort after Huang departure
Hugging Face CEO Clément Delangue said the company would keep fighting for open-source AI after Nvidia CEO Jensen Huang’s departure. Huang had supported open-source AI and open-weight models, raising questions about Nvidia’s approach under new leadership. Nvidia’s planned acquisition of Hugging Face remains scheduled to close in the first half of 2027, while the U.S. Department of Justice has made inquiries about the deal’s effect on competition. Huang has argued that open models widen access, strengthen competition and give customers more control. Nvidia also announced a $6 billion deal with coding startup Poolside in August to support an open AI ecosystem in the U.S.
Costco forecast got the right answer for the wrong reasons
Hassan El Mghari turned 214 Costco filings into 3,000 timestamped facts and tested GPT-OSS-120B with 16 runs costing 19 cents. Five runs passed his self-correction checks and predicted a $560 million profit increase over two years. His father, auditor Omar El Mghari, said two flawed assumptions canceled each other out: the model counted a membership price increase after July 2024, while using a 15% estimate for executive memberships instead of the disclosed 45%. Correcting the estimate changed the forecast error from 1.7% to 28%. Hassan said the test method was still valid because GPT-OSS-120B and Llama-3.3 70B did not know about the price increase, unlike GLM-5.3 and Chimera-3.
Google Gemini flaw let users download images its filters blocked
A technical flaw in Google Gemini let users download images that the AI’s security filters had blocked. The images could include potentially harmful or copyrighted material, such as Mickey Mouse depictions and unauthorized photo edits. Researcher Jason Kao of cybersecurity firm Fog Security found a gap between Gemini’s chat interface and Google’s internal file delivery system. The issue showed that protections could fail in the systems delivering files, even without users tricking the AI or changing its instructions.
BDO grows its AI tools by helping employees build them
BDO says it is treating AI as a people transformation, not just a technology purchase. In an anonymous survey in April and May, nearly 3,500 colleagues responded, and 87% said they already used AI; they also submitted more than 11,000 ideas for using it at work. Before its first formal training, more than 600 team members had built over 1,200 agents, and training AI champions helped the number of agents rise by about 80% in two weeks. One employee, Emma, used AI to update nearly 2,000 design files in about a day, instead of a 20-person effort that took close to a month. BDO says leaders set rules for AI use, track business impact, and scale tools that work while keeping human judgment in place.
Sources
- Wix Stock Is Up 6.2% What's Changed: AI Push Deepens
- Is CrowdStrike Holdings (CRWD) Redefining AI Security With Its New Agent Tools?
- How Investors Are Reacting To CrowdStrike Stock After New AI Security Launches
- Alphabet vs. Broadcom: the best AI stocks for risk and reward
- Alphabet vs. Broadcom: the best AI stocks for risk and reward
- Sakana AI hires US spies to sell defense
- AI is exposing the impact problem: 3 questions to ask before any learning investment
- AI's long list of problems just for a whole lot worse
- A Worker Refused to Use AI on Religious Grounds. His Employer Said Yes
- Hugging Face CEO vows to keep fighting for open-source AI after Jensen Huang's exit
- Costco forecast was right for wrong reasons
- Google's secret AI bypass: How Gemini sneaked out banned images
- The best AI lab in business is not for sale—you already own it
Comments
Please log in to post a comment.