Tether released a 254-million-parameter speech model that moves transcription onto phones, laptops, smart glasses, and robots instead of sending audio to the cloud. The Tether TranscribePsy release targets a difficult balance: keeping a model small enough for consumer hardware without accepting the usual accuracy loss. The company says TranscribePsy runs faster than every competing model it tested across six smartphones. It also reports a 6.58 percent word error rate on Multi

Vulcan Infrastructure and Power faces a local moratorium campaign as residents question a reported 260-megawatt grid expansion tied to its Torrey property. The proposed Vulcan AI data center moratorium would temporarily block new or expanded computing facilities while the town reviews its zoning rules. The dispute is not simply another fight over a new industrial development. Vulcan's Dresden site already combines a natural gas power plant, electrical infrastructure, water pe

Dapta raised $6 million to expand its AI sales agents, putting fresh capital behind an unresolved small-business question: can automation convert more leads into paying customers? The Dapta $6M seed round gives the company resources to pursue that question across the United States and Latin America. Dapta says its technology answers inquiries, qualifies prospects, schedules appointments, and follows up through voice and text channels. That pitch challenges a familiar sales mo

TypeSafe AI shipped the Jev 1.14.0 release on October 14 with images, bounding boxes, and longer context, despite labeling those capabilities unsafe for production. That tension is the real story. TypeSafe built Jev around constrained, typed decisions instead of free-form generation. Now it is letting the model inspect images and locate objects, while warning that the results remain too uneven for dependable automation. The company calls this opt-in layer its “danger zone.” D

Intel has expanded its Intel EMIB and Foveros roadmap despite one stubborn constraint: sophisticated packaging only matters when factories can deliver it reliably. The company now envisions packages exceeding 12 reticles, carrying more than 24 high-bandwidth memory stacks, and combining horizontal bridges with dense vertical bonds. That plan changes how Intel's two best-known packaging technologies fit together. EMIB once handled lateral connections, while Foveros stacked die

Floqer AI agents now research sales prospects across more than 100 data sources, despite a persistent problem: automated research is only useful when teams trust it. The startup says its system can find accounts, identify contacts, enrich records, and monitor changes without constant manual searches. That proposition targets one of revenue operations’ least glamorous jobs. Sales teams routinely move between Google, LinkedIn, company websites, job boards, data vendors, call tr

The Philadelphia Semiconductor Index selloff erased about $1.6 trillion in market value during six consecutive losing sessions. The decline challenged the assumption that accelerating artificial intelligence investment would keep supporting every major chip supplier. The closely watched index, commonly called the SOX, lost roughly 10% during the streak. Its October 14 decline extended the longest losing run since May 2025, according to the six-session rout initially reported

Google released its geospatial reasoning agent after four benchmark evaluations, moving Earth AI from isolated models toward automated, multi-step spatial analysis. The experimental system connects Gemini with specialized models for weather, floods, population patterns, and remote sensing. It also searches large data catalogs, writes geospatial code, and trains task-specific predictive models. Google is expanding early access through Ask Google Earth, cloud workflows, and an

Elon Musk says the Grok Voice 3.0 benchmark result makes xAI’s model the world’s leading conversational voice system. Artificial Analysis does place it first, with an overall Speech-to-Speech Index score of 86.9. That result puts Grok Voice 3.0 ahead of GPT-Live-1 with Astra at 83.2 and Gemini 3.8 Live at 82.6. It also represents a 5.6-point increase over Grok Voice 2.0, which scored 81.3. The headline is accurate within the boundaries of this benchmark. Yet the underlying re

Claude Code idle compaction can now summarize a large session before its prompt cache expires, even when the context window is not full. The feature targets sessions containing at least 200,000 tokens by default. Its central tradeoff is stark: Claude returns to a cheaper, shorter context, but continues from a generated summary instead of every original turn. Anthropic engineer Lydia Hallie described the behavior in an October 2026 idle compaction post. Users can disable the f

Google is ending Gemini AI Plus Pro model access on November 13, 2026, despite offering that capability when the subscription expanded earlier this year. After the cutoff, affected subscribers will retain Flash and Flash-Lite but lose the option to select Pro. The change applies to both the Gemini app and AI Mode in Google Search, according to a subscriber email reported by 9to5Google. Existing conversations created with Pro can continue, but users must switch those chats to

McKinsey is giving some entry-level applicants 10 minutes with a simulated customer, turning the McKinsey AI client interview into a test of live judgment. The avatar plays a business leader while human assessors watch how each candidate listens, responds, and adjusts. The pilot is notable because the software reportedly does not score applicants or decide who advances. AI supplies the unpredictable conversation, but McKinsey partners and assessors retain responsibility for e

The IMF AI financial stability report has moved the AI bubble debate beyond technology stocks. Its warning is direct: an AI market correction could spread through debt, concentrated portfolios, private credit, and tightly connected corporate balance sheets. The concern is not simply that Nvidia, Microsoft, or another prominent company might lose market value. The deeper risk is that AI investment now supports economic growth, corporate borrowing, data center construction, and

DeepL is reporting surging demand for spoken translation, but DeepL voice AI growth now brings the company into more direct competition with OpenAI and Anthropic. CEO Jarek Kutylowski discussed that pressure with Bloomberg’s Tom Mackenzie at the International AI Summit in Dublin on October 14. The timing matters. DeepL has expanded voice-to-voice translation, integrated technology from an acquisition, adopted AWS infrastructure, and cut about one-quarter of its workforce duri

Zenzap funding has reached $44 million as the startup challenges WhatsApp with an AI workplace messenger designed for frontline teams. Its pitch is simple but ambitious. Work conversations should feel like ordinary mobile chat without leaving company data inside employees’ personal accounts. The company is targeting a persistent gap between consumer messaging and enterprise collaboration software. WhatsApp feels familiar to employees, while Slack and Microsoft Teams offer adm

Samsung is extending four practical Galaxy AI tools to 24 languages, despite leaving users without a firm date for receiving the new support. The Samsung Galaxy AI language expansion adds Hungarian and Malay to Call Assist, Note Assist, Interpreter, and Transcript Assist. These features translate calls, organize notes, interpret conversations, and convert recordings into searchable text. Samsung also plans a broader language update for Now Nudge, its system for suggesting act

Tesla SpaceX merger speculation gained a specific deadline on October 14, when Barclays analysts pointed to Tesla’s approaching 2026 annual meeting. Their forecast puts a possible announcement before December 6, despite the absence of a disclosed agreement, exchange ratio, or shareholder proposal. Analysts led by Dan Levy argue that the strongest rationale lies in artificial intelligence, not ordinary financial savings. Tesla brings vehicles, robots, energy storage, manufactu

YouTube AI search is moving beyond Premium, with Ask YouTube rolling out to U.S. users across Android and iOS. The change turns a restricted experiment into a mainstream mobile feature, although availability will expand gradually. That shift matters because Ask YouTube does more than place a chatbot beside conventional results. It can interpret a complex request, combine videos with written information, identify relevant clips, and continue the search through follow-up questi

Alibaba launched its new AI glasses in China on October 13, giving Qwen sharper cameras, eye tracking, payments, and deeper access to everyday services. The Qwen Glasses N1 series moves Alibaba beyond a simple model demonstration. It puts the company’s AI assistant inside a device that can see, hear, respond, and act while users move through the physical world. The launch also establishes a clearer contest with Meta. Both companies want glasses to become a regular entry point

Apple has introduced its fourth-generation Apple TV 4K, turning a familiar streaming box into a hands-free Siri AI hub with three built-in microphones. The update is more significant than a routine processor refresh. Apple is combining conversational entertainment search, smart home controls, camera summaries, and a new four-speaker HomePod setup in one television interface. That strategy places the Apple TV 4K Siri AI experience in a broader contest with Amazon and Google. T