- AI Tangle
- Posts
- ☕️ Anthropic Brings Claude 3 to Mobile
☕️ Anthropic Brings Claude 3 to Mobile
The AI world remains as busy as ever with names like Anthropic releasing an iOS app to access its flagship Claude 3 along with premium business plans and Microsoft continuing its journey through Southeast Asia to build up its AI infrastructure. Meanwhile, Amazon says goodbye to CodeWhisperer and hello to Q Developer and Airbnb is trying out some new group booking features and AI-enhanced customer support. Join us at AI Tangle as we untangle this week's happenings in AI.
THE BIG AI STORY
Anthropic, the company behind the mighty Claude 3 family of generative models consisting of Opus, Sonnet, and Haiku, has recently debuted an iOS app that allows users to bring Claude 3 with them wherever they go more conveniently than a web app. To add to it, Anthropic is also releasing a new premium business plan called Teams with many added features to allow companies to get more out of Claude 3.
What are the details?
The iOS app shouldn't come as a surprise after Anthropic had been digging around to hire iOS developers in the past few months, syncing with the web client and tapping into Claude 3’s vision capabilities to offer real-time analysis for uploaded and saved images. The new Teams business plan gives users a bigger context window (200,000 tokens) compared to the company's Pro plan and more toggles to control billing and user management. In the coming weeks, it will also gain many collaboration features like integration with codebases and customer platforms and a canvas to work with team members on AI-generated docs and projects.
6 QUICK HITS
Amazon's fairly successful AI-powered coding assistant was recently rebranded as Q Developer, making it part of the company's Q family of business-oriented AI chatbots. Amazon's Q Developer, much similar to the ever-popular GitHub Copilot, can generate and test code and autonomously perform tasks like implementing features and refactoring existing code. Q Developer is available both for free, though with some limitations, and as a $19 monthly subscription.
Following Microsoft CEO, Satya Nadella, visit to Indonesia earlier this week to elaborate on the country's cloud and AI infrastructure, the company has now set in motion a plan to build its first regional data center in Thailand to expand accessibility of its cloud services. The Thailand data center will also help enterprise-grade reliability, performance, and compliance with data residency and privacy standards, Microsoft added in the statement.
Paul Dean, a State Department arms control official, recently stated in an online briefing the importance of only ever allowing humans, not AI, to make the final decision on the usage of nuclear weapons. Dean's message is directed towards China and Russia and urges the two superpower countries to follow in the steps of the United States, a country Dean claims is leading the way in terms of AI governance and regulation in military affairs.
Microsoft to disallow U.S. police departments from using the company's models for facial recognition
Recently, tech giant Microsoft updated its terms of service to explicitly disallow the use of the company's Azure OpenAI Service AI for facial recognition by U.S. police departments. This could follow Axon's, a manufacturer of tech and weapons, recent product release that leverages OpenAI's GPT-4 to summarize audio from body cameras, as critics were quick to point out the dangers of AI's well-known hallucination and racial bias problems.
For its annual summer update, Airbnb is introducing a few updates for group booking and a new category called "Icons," experiences hosted by celebrated names in the media. The group booking feature allows people to create shared wishlists, invite others to join, and vote on properties. In addition, Airbnb is introducing a messaging tab where attendees can directly message the host, who will be able to use AI-powered suggestions to reply to questions and the like.
Israeli-based AI security company Apex emerges with $7 million in seed funding, backed by Sam Altman
Apex, an Israeli startup specializing in AI security, has raised $7 million in a seed funding round co-led by Sequoia Capital and Index Ventures, with additional backing from the poster child and CEO of OpenAI, Sam Altman. The company's platform provides real-time monitoring of AI usage, threat detection, and rapid response to security incidents. The recently bagged $7 million in funding will go towards accelerating product development, expanding the team, and increasing go-to-market efforts.
4 AI TOOLS
Streak - AI-powered data entry, precise insights, and tailored suggestions - Streak AI is your team's CRM co-pilot designed to help your team make informed decisions.
Lume - Revolutionize your workflow with Lume by allowing you to map and create data pipelines in seconds, then control them with Lume's AI functionality to review, edit, and deploy them indefinitely.
Outerbase - Navigate your data with AI - View, edit, query, and visualize your data in one powerful and easy-to-use interface tool for your database.
Collato - Collato is an AI meeting assistant and notetaker that transforms meetings into easy-to-read documents with ease, allowing users to focus on the meeting, rather than jotting down notes.
AI READ & WATCH
It's Just an Android App (3-min read)
The first reviews for the Rabbit R1 aren't looking pretty, and jokes back from CES 2024 earlier this year asking why it couldn't have just "been an Android app" are resurfacing, and for good reason. Among many other complaints with the gadget, it's not looking good.
The Stealth Reveal of... GPT2? (27-min watch)
Earlier this week, people took notice of a curious new model on the Large Model Systems Organization (LMSYS) website, dubbed GPT2, not to be confused with GPT-2, a wildly outdated model by today's standards. Wes Roth, a YouTube channel covering the latest AI developments, looks at the stealth reveal.