Gemini Omni is Google DeepMind's video generation line. Google (@Google) posted the 1.2 Flash update for AI Studio and related products. Flash models are Google's faster, lower cost tier.

Google (@Google) posted on Nov 5 introducing Gemini Omni 1.2 Flash, an updated video generation model from Google DeepMind.

According to Google, the model offers stronger consistency, more creative control and faster, more cost effective video at scale. Google said users can draft at 360p or 720p for $0.07 per second and upscale to 1080p or 4K, for about 30% lower cost. Clips can run up to 15 seconds, or any exact length from 3 to 15 seconds.

Google also said users can edit in conversation, swapping backgrounds, subjects or lighting turn by turn without starting over, get more natural multi speaker dialogue with clearer lip sync, and direct continuous camera moves with fewer jump cuts.

These figures and feature claims come from Google. The post does not compare Omni 1.2 Flash against rival video models by name.

The update lands in a crowded race for AI video, where length, cost and controllable editing decide whether creators use a model for drafts or finished work.

The post passed about 270,000 views.

Watch for creators to test the 15 second clips and conversational edits against Runway, OpenAI and other video tools. Pricing at $0.07 per second for drafts will shape who uses it at volume. Google may follow with higher quality tiers or longer clips. Any reports of lip sync or consistency failures will also matter. Studio integrations and partner apps will show whether Omni 1.2 Flash becomes a default draft tool.

Gavin Baker (@GavinSBaker) is an investor who covers AI infrastructure. Gina Hinojosa is the governor elect of Texas after the Nov 3 elections. Data center permits and power rules became a midterm issue in several states.

Gavin Baker (@GavinSBaker) posted on Nov 5 that Texas governor elect Gina Hinojosa has explicitly said that all currently permitted data centers in Texas will be allowed to finish construction, covering all of 2027 and likely well into 2028.

According to Baker, she plans laws so newly permitted data centers for delivery in 2028 and 2029 will pay taxes, commit to no increase in power prices for consumers, avoid hurting water availability, use union labor and give Texans a seat at the table. He called that reasonable and said she does not want to lose blue collar jobs and property taxes for small towns. He pointed to an interview segment starting at 6:30 and said he would attach her campaign position next.

These points relay Baker's reading of her remarks and campaign materials. They are not a signed statute.

The post is a concrete counterweight to fears that Democratic wins would freeze data center buildouts overnight. Texas alone is often cited as a large share of planned US capacity, including in a Goldman Sachs warning already in the Nov 4 deck.

The post drew more than 4.5 million views.

Watch for Hinojosa's official statements and any early bills on taxes, power prices, water and labor for new data centers. Developers with Texas permits will update project timelines. Other states may copy a permitted projects finish, new projects face stricter rules split. AI labs and cloud companies will weigh Texas against other sites. Campaign site language versus signed bills will decide how firm the finish permitted projects promise really is.

Walter Bloomberg (@DeItaone) is a markets headline account. Google has been expanding data center sites across the US to support Search, YouTube and AI services including Gemini.

Walter Bloomberg (@DeItaone) posted on Nov 5 a headline stating that Google is building its first data center in Wyoming, representing a $17 billion investment.

The post is a short headline relay and does not link a Google press release or give the site town, power source, timeline or job count in its text. Those details should be checked against Google's own announcement.

Even as a headline, a $17 billion first site in a new state is a large commitment. It lands on the same day as other data center stories in this deck, including Texas permit signals, a reported dump of data center loans, and talk that OpenAI leaders may lower capacity expectations after the elections.

If the figure holds, it shows at least one major cloud company still planning big buildouts even while markets sell off AI names and local politics turn cooler on data centers.

The post passed about 170,000 views.

Watch for Google's official release with location, megawatts, water use and opening date. Local officials in Wyoming may comment on taxes and jobs. Other hyperscalers' site announcements will show whether big spending continues after the midterms. Any force majeure or delay notices on other projects would contrast with this investment. Grid and water permits in Wyoming will also show how fast the site can actually energize.

Walter Bloomberg (@DeItaone) relays company and market headlines. Claude is Anthropic's assistant. Pro, Max and Ultra are paid Claude subscription tiers.

Walter Bloomberg (@DeItaone) posted on Nov 5 that Anthropic is introducing Personal Finance in Claude, with a US beta rollout beginning this week for Pro, Max and Ultra subscribers.

According to the post, the feature provides read only access to financial information and is not designed to offer investment advice. Anthropic also confirms no ads or sponsored recommendations, the post says.

These details are secondhand from the headline account and should be checked against Anthropic's own product post or docs. The post does not name which banks or brokerages connect, or how account linking works.

The launch fits a wider move of assistants into money. OpenAI has been adding finance voice features in ChatGPT, and users already ask models about budgets and bills. A read only design with a no advice framing is Anthropic's way of entering that market while limiting regulatory risk.

ChatGPT and other assistants have already edged into budgets and bill questions, so Anthropic is responding to a feature users now expect in a paid assistant.

The post passed about 195,000 views.

Watch for Anthropic's official product page listing supported institutions and permissions. Bank partners and security reviews will decide how useful the beta feels. Competing assistants may match the feature. Any shift from read only into advice would draw regulator attention. User trust will hinge on clear permissions screens and on whether balances and transactions stay accurate day to day. Early reviews will also say whether the US only beta is enough for people with accounts at big banks.

Walter Bloomberg (@DeItaone) relays a CNBC report. Moonshot AI is the Chinese lab behind Kimi models, including Kimi K3.1 Pro whose weights were released on Nov 4. US rules limit shipping advanced Nvidia chips to China.

Walter Bloomberg (@DeItaone) posted on Nov 5 that China's Moonshot AI has reportedly explored renting Nvidia's advanced Blackwell computing capacity in Europe, potentially getting around US limits on chip exports to China.

According to the post, which cites CNBC, talks included access to clusters of 8,000 or more Blackwell GPUs, with Moonshot potentially offering 30% to 50% higher upfront payments than customers outside China. The post says US export controls restrict physical chip shipments to China, but remote access through overseas cloud providers remains a regulatory gap, and Washington is considering legislation to close that loophole.

These claims are secondhand from CNBC via Walter Bloomberg. They are not confirmed by Moonshot or Nvidia in the post.

The story matters because open weight releases from Chinese labs already compete with US models. Cheap remote access to Blackwell would widen that race even if boards never enter China.

The post passed about 130,000 views.

Watch for the full CNBC report, and for any comment from Nvidia, Moonshot or European cloud providers. US lawmakers may move bills on remote access. Other Chinese labs' cloud deals in Europe or the Middle East would show how wide the gap is. Training claims that need large Blackwell clusters will be read against this report. Allied governments may also tighten rules on who can rent advanced GPUs without an export license.

unusual_whales (@unusual_whales) relays markets news. Stargate is the large data center project tied to OpenAI, SoftBank and Oracle. Force majeure is a contract claim that events beyond a party's control may excuse performance.

unusual_whales (@unusual_whales) posted on Nov 5 that a mystery seller is trying to dump about $500 million in data center loans all at once, according to Bloomberg, days after the midterms raised doubts about many projects.

According to the post, the anonymous seller is auctioning a dozen loans, including a $25 million piece of financing for a huge Stargate site in New Mexico, the joint project of OpenAI, SoftBank and Oracle. The post says Oracle has already sent that site's developer a force majeure notice, a legal claim that events beyond its control may excuse it from its obligations.

These details are secondhand from Bloomberg via unusual_whales. The seller is unnamed in the post.

A forced sale of this size would show lenders or investors trying to exit AI infrastructure credit just as politics and power costs get harder. The Stargate mention ties the loan dump to one of the highest profile AI buildouts.

The post passed about 880,000 views.

Watch for Bloomberg's full story naming more loans and the seller type, and for Oracle or the New Mexico developer to comment on the force majeure notice. Bid results on the auction will show how much credit investors still want data center risk. Other project delays or notices would extend the signal. Credit spreads on data center ABS and loans will show whether the dump was a one off or the start of a wider exit.

unusual_whales (@unusual_whales) relays a Wall Street Journal report. OpenAI's data center needs are a major driver of AI chip and power demand. Goldman Sachs' warning that Democratic wins could affect 45% of planned US capacity is already in the Nov 4 deck.

unusual_whales (@unusual_whales) posted on Nov 5 that OpenAI's leadership is already weighing the fallout from Tuesday's elections.

According to the post, by Wednesday morning senior leaders were discussing whether the results meant OpenAI would have to lower expectations for its data center capacity next year, per the WSJ, citing a person familiar with the conversation. The post adds that voters in at least 29 places across the US approved bans or restrictions on data centers on Tuesday, and that Goldman Sachs has warned Democratic wins at the state and local level could put 45% of planned US data center capacity at risk.

These points are secondhand from the WSJ via unusual_whales. OpenAI has not confirmed the internal discussion in the post.

The report sits beside other Nov 5 signals: Texas saying permitted projects can finish, a $500 million loan dump, and Google's reported Wyoming build. Together they show the post election data center picture is mixed, not one way.

The post passed about 200,000 views.

Watch for the full WSJ story and any OpenAI comment on 2027 capacity plans. Chip and power suppliers' guidance may shift if OpenAI softens demand. Local ban counts and court challenges will decide how binding the 29 place figure is. Goldman's 45% risk estimate will be tested against actual permit outcomes. Supplier comments on OpenAI related orders would confirm whether internal talk turned into slower build plans.

The Kobeissi Letter (@KobeissiLetter) is a markets newsletter account. The Nov 5 open followed the midterms and a multi day slide in AI related stocks.

The Kobeissi Letter (@KobeissiLetter) posted on Nov 5 that US technology stocks collapsed at the open, erasing $1.3 trillion in market cap in 30 minutes, with dozens of names down 10% or more.

The post lists moves it says include SanDisk down 15%, Western Digital and Nebius down 13%, AMD down 12%, Seagate, Qualcomm, Micron, Tesla, Dell, Arm, Marvell and CrowdStrike around 10%, Intel down 9%, CoreWeave and Palantir down 8%, SpaceX and TSMC down 7%, and Nvidia, Broadcom and Oracle down 5%. It says the Nasdaq 100 was down 4% on the day and that a highly eventful day lay ahead.

These figures are from Kobeissi at the time of the post and can change as the session continues. They are not official exchange settlement numbers.

Many of the hardest hit names are tied to AI infrastructure, chips, storage and data centers, so the tape is a market verdict on the AI trade after the elections.

The post passed about 840,000 views.

Watch the close and whether AI chip and data center names keep falling in the following sessions. Lab funding rounds and IPO talk may cool if the tape stays weak. Earnings guidance from Nvidia, AMD and Oracle will be read against this move. Political news on data centers can keep adding volatility. Options pricing on Nvidia and other AI names will show how long traders expect the shock to last.

unusual_whales (@unusual_whales) cites pre IPO perpetual prices on Liquid (@liquidtrading). Implied valuations on such platforms are not the same as a completed IPO price.

unusual_whales (@unusual_whales) posted on Nov 5 that Anthropic and OpenAI have seen more than $700 billion erased from their combined implied valuations over the past month as the selloff in AI markets continues.

According to the post, Anthropic has fallen nearly 18% to $1.738 trillion, while OpenAI is down over 18% to $1.370 trillion, based on pre IPO perpetual prices on Liquid.

These numbers come from a secondary trading venue via unusual_whales. They are not official company valuations and can differ from primary funding rounds.

The charts attached show Anthropic's Liquid price around $1,738.70, down about 18.7% over 30 days, and OpenAI around $1,393.70, down about 17.8% over the same window. The OpenAI figure in the image is close to but not identical to the $1.370 trillion line in the text, so both should be treated as platform marks.

Even as rough marks, a combined $700 billion drawdown in a month matches the public tech selloff and the weaker Polymarket odds on Anthropic's IPO size.

The post passed about 260,000 views.

Watch whether Liquid and other pre IPO venues keep marking the labs lower, and how that compares with their next primary funding rounds. Public AI chip stocks will influence the private marks. Any IPO timeline comments from Anthropic or OpenAI will be read against these prices. Large secondary trades reported by brokers would confirm whether the Liquid marks match real deals.

Apoorv Shankar (@lazyapprove) is launching Aina. The post describes a ring with a small screen that shows suggested actions from apps and agents for quick approval.

Apoorv Shankar (@lazyapprove) posted on Nov 5 introducing Aina, a ring with a screen that he says predicts and shows the three things you will need next, so one flick of your thumb can approve them.

According to the post, AI is already proactive but still needs a yes, which today means unlocking a phone, opening an app and tapping approve many times a day. Aina is pitched as AI's own interface. Features listed include Cues for proactive suggestions, Voice to talk to agents and take notes, Remote to control music, Mac and TV, and Calls from the hand. Pre orders are open, the post says.

These product claims come from the maker. The post does not give price, ship date or which agents are supported at launch.

The idea is related to other agent hardware demos this week, including Razer's GrokPuck, but Aina focuses on tiny, always on approvals rather than a full pocket assistant screen.

The post drew about 2.6 million views.

Watch for price, ship dates and which agent platforms support Aina at launch. Reviews of battery life, false suggestions and privacy will decide whether a ring beats phone prompts. Phone and watch makers may add similar one flick approve flows. Competing rings and pins may appear quickly if pre orders are strong. App store listings and agent SDK support will show whether Aina is a real platform or a single purpose gadget.

Min Liang Tan is the CEO of Razer. He posted early on Nov 5 Eastern that his team hacked together GrokPuck, described as Grok Bot in your pocket, with a swipe to change the bot. Elon Musk quote posted and promoted Grok Bot repeatedly the same morning, including a post that quoted the GrokPuck demo. Phone calling demos for Grok Bot already appeared in the Nov 3 deck.

Min Liang Tan (@minliangtan), CEO of Razer, posted just after midnight Eastern on Nov 5 that he had been using Grok Bot and that his team hacked together GrokPuck, calling it Grok Bot in your pocket. The video shows a small device where swiping the screen changes the bot.

Elon Musk (@elonmusk) quote posted the demo with "Everyone is using Grok Bot" and separately told followers to try Grok Bot, quoting other user demos. Those amplify posts pushed the GrokPuck video past two million views within hours.

This is a prototype demo from a hardware executive, not an official Razer product launch or an xAI hardware announcement. It sits in a week when Grok Bot phone calling and booking demos were already circulating. The story is the sudden jump from software agent demos into a pocket form factor people can film and share.

Shubham Saboo and others also posted that they were consolidating personal agents onto Grok Bot the same day cycle, which adds to the product buzz without being a separate launch.

Watch for whether Razer or xAI productizes a GrokPuck style device, and for more hardware makers to film similar agent gadgets. Compare retention and real tasks against phone based Grok Bot, OpenAI and Anthropic agents. If it stays a one off hack video, the lasting signal is still how quickly Grok Bot demos travel when Elon amplifies them. A real product SKU from Razer or xAI would turn this from a viral demo into a hardware story.

Elon Musk quote posted Congresswoman Yassamin Ansari's Nov 4 call for an AI pause, industry regulation and a ban on artificial superintelligence. Ansari's original post is covered in the Nov 4 deck. Musk's line is a short reaction, not a policy proposal of his own.

Elon Musk (@elonmusk) posted early on Nov 5 Eastern, "It's a Brave New World," quoting Congresswoman Yassamin Ansari's call for a pause on advanced AI, regulation of the industry and a ban on artificial superintelligence.

Ansari's original post is already in the Nov 4 deck and had crossed several million views. Musk's quote pushed the exchange back into the morning feed with nearly a million views of its own within hours.

The post is a cultural and political reaction, not a new bill or company action. It matters because Musk's amplification often sets the tone of the day's AI policy argument on X, especially right after an election that already featured AI pause talk in Michigan and elsewhere.

Readers should open Ansari's own post for the agenda language and treat Musk's line as commentary that made the fight travel farther, not as a substitute for the agenda language she wrote herself.

Watch whether other lab leaders or members of Congress reply in kind, and whether Ansari or allies turn the day 1 agenda into introduced legislation. The pairing of her post and Musk's quote will likely keep showing up in coverage of post election AI politics through the week. Watch also for rival lab CEOs to answer with their own framing of pause talk and ASI bans.