{{first_name | Hey}}, welcome back.
Today, Claude makes voice AI more useful on the go, Google introduces a model designed to tackle cybersecurity threats, and Salesforce prepares to charge for activity from external AI agents. Plus, Reddit says declining search traffic is pushing it toward a more login-focused experience.
These are today's updates.
Stop letting messages slow down decisions with Wispr Flow*
Claude’s voice mode gets an upgrade
Google’s new AI model takes the lead on a major benchmark
Salesforce plans to charge external AI agents
Tools, jobs, resources, and last issue’s winning prompt ⬇️
Together with Wispr Flow
Your workday is full of things to say, but getting them onto the screen means typing every word.
Wispr Flow turns your natural speech into clean, formatted text that's ready to send, in any app on any device. Speak as you would in a conversation, change your mind mid-sentence, or ramble through a complicated thought. It removes filler words, handles corrections, and takes care of punctuation.
Use it to reply to messages, draft emails, or capture ideas before they disappear. No integrations or complicated setup required.
Wispr Flow works on Mac, Windows, iPhone, and Android, and is much faster than typing. Use the link below to get one month of Wispr Flow Pro for free with unlimited words.
*Sponsored and dictated by Wispr Flow
Top News
Anthropic updated the Claude mobile apps. Voice mode on iPhone now works with Apple CarPlay, so you can ask Claude something on a drive with your hands on the wheel. According to Claude Voice mode on iOS and Android now answers sooner and hears you better in a noisy room. The apps can also stay signed into several accounts at once.
Claude is available in 33 languages, including new ones like Bengali, Swahili, Thai and Turkish. These new features are a delight for commuters, anyone who juggles work and personal accounts, since voice now works hands-free and switching no longer needs a sign-out. Pull-down search for iPhone and iPad is coming soon too. Check out how to use it.
Google announced Gemini 4 Carbon, which ranks first on version 5 of the Artificial Analysis Intelligence Index with a score of 59, ahead of Claude Opus 5.5 at 58 and GPT-6 Astra at 57. Access starts this week for selected cyber defenders through Google’s Fairwind Program, with standard pricing of $4 per million input tokens and $20 per million output.
This one’s for security teams, since Google reports leads in finding, exploiting and patching vulnerabilities. Carbon was tested at its high reasoning setting while Opus and Astra ran at maximum. Google hasn’t set a date for developers or consumers. Explore Gemini Carbon 4 here.
Come November, Salesforce is launching agent registration and a new billing model for outside AI agents, such as Claude. It has released the details in its help article. Every successful call a registered agent makes, over MCP or an API, will use Flex Credits, which Salesforce sells at $500 per 100,000. The rate per call is still listed as TBA, while metering starts only after 30 days’ notice.
Customers who bought after Sept 17 must register agents within three months of Salesforce’s go-ahead. Existing customers using APIs need to migrate at renewal, and anyone using a Salesforce MCP server must register regardless. Note that sandbox and developer orgs aren’t metered. For more information click here and to read Salesforce’s announcement click here.
Don’t miss this: Reddit says Google’s AI Mode sends it no traffic, and it is moving toward a login wall. Its search referrals fell by more than 1 million a week on average in October. Reddit warned that its daily users could fall by a low double-digit percentage next quarter if the trend continues. Its shares dropped about 18% after hours. Read the report here.
Signals
Tools
Imagine AI V2: An agent that runs LinkedIn for the founder’s and his team
Floot in ChatGPT: Build apps in ChatGPT; Publish it in minutes
WordCraft: Open-source word processor; Free alternative to MS Word
Resources
Poll
Choose a prompt you want us to publish next
Market
Funding
FuriosaAI raised $620M to scale efficient, production-ready AI inference
Upwind raised $400M to build the cloud and AI security platform
Roles in AI
Brand Paid Media Manager at Cursor
Tutorial
Think Claude Code is only for developers? Vaibhav Sisinty’s beginner-friendly tutorial walks through 14 features that can help you automate research, create on-brand presentations, schedule daily briefings, do deep research, and manage multiple AI tasks, and the 7 slash commands - all without being an engineer.
Everything we don't have room for here, we post on X. Go see for yourself
Prompt of the Day
AI Budget Review Assistant
When to use this?
Use this prompt before budget reviews, quarterly planning, or AI investment decisions to spot cost overruns, uncover savings opportunities, and prioritize spending based on business value.
Act as an AI Budget Review Assistant helping enterprise AI leaders and mid-level executives review budgets, control costs, and make informed resource allocation decisions.
Your job is to analyze the budget information...
{Click on the button to view full prompt}Get more such prompts in the Prompting Playbook (free for you)
Stay curious, {{first_name | reader}}
PS. If you missed yesterday’s issue, you can find it here.
*sponsored content
