Good Morning AI followers!

It’s been a big week for Alphabet! This week’s main stories are about Alphabet having revised its AI principles, allowing AI use in weapons and surveillance, citing national security needs and sparking ethical debates and another one about Google unveiling Gemini 2.0, a collection of AI models designed for various tasks and budgets, advancing the accessibility and versatility of AI agents.

This week’s edition:

• Google Parent Company Alphabet Revises AI Ethics, Opens Door to AI Weaponry

• Gemini 2.0 Unleashed: Google's AI Agent Revolution Begins

• Your 5-Min AI Learning: Search for the Needle in Your Bookmark Haystack with mindECHO

• New AI Tools

• More AI News

• AI Image of the Week

Read time: 8 minutes

LATEST NEWS

ALPHABET ALTERS AI ETHICS

Google Parent Company Alphabet Revises AI Ethics, Opens Door to AI Weaponry

What You Need to Know

Alphabet, Google's parent company, has altered its AI principles, removing the explicit prohibition of AI being used for weapons development and surveillance tools. This decision marks a shift from Google's previous stance against AI applications that could cause overall harm. The company defends this change by citing the need for collaboration between businesses and democratic governments to develop AI that protects national security. Alphabet's executives, Demis Hassabis and James Manyika, argue that democracies should lead in AI development, guided by values like freedom and human rights.

The Finer Details

• The revised AI guidelines eliminate the clause that previously restricted AI use in applications "likely to cause harm".

• The company believes that AI principles should evolve alongside the technology to reflect its varied applications.

• Alphabet plans to invest approximately $75 billion (approximately £59 billion)  in capital expenditures in 2025 to expand its AI capabilities and infrastructure.

• The shift in policy aligns with the broader industry trend of recognising AI's potential in national defence.

• The announcement coincided with Alphabet's latest financial report, which fell short of market expectations, causing a stock price dip.

So What?

The decision sparks debate over the ethics of AI in warfare and surveillance, as well as the influence of commercial interests on AI development. Concerns remain about AI weapons making autonomous decisions on the battlefield. This shift comes after previous internal protests within Google against AI military contracts, highlighting the ongoing ethical considerations surrounding AI's role in society.

GEMINI 2.0 FOR ALL

Gemini 2.0 Unleashed: Google's AI Agent Revolution Begins

What You Need to Know

Google has made its Gemini 2.0 AI model suite available to the general public, marking a significant step in the development and accessibility of AI agents. This move is part of Google's broader strategy to invest heavily in AI agents, as it competes with other tech giants like Meta, Amazon, Microsoft, OpenAI, and Anthropic in the AI arena. Gemini 2.0 includes models like 2.0 Flash, 2.0 Pro Experimental, and 2.0 Flash-Lite, each designed for different tasks and cost considerations.

The Finer Details

• Gemini 2.0 Flash: Optimised for high-volume, high-frequency tasks at scale.

• Gemini 2.0 Pro Experimental: Designed for enhanced coding performance.

• Gemini 2.0 Flash-Lite: Google's most cost-efficient model.

• Cost: Gemini Flash costs developers 10 cents per million tokens for text, image, and video inputs, while Flash-Lite costs 0.75 cents for the same.

• AI Agents: These models can understand the world, think ahead, and take action with user supervision, featuring advances in multimodality and native tool use.

• Competition: Companies like Anthropic and OpenAI are also developing AI agents with similar capabilities, such as using computers like humans to complete complex tasks.

So What? 

Google's release of Gemini 2.0 to the public signifies a major push towards making AI agents more accessible and versatile. By offering a suite of models tailored to different needs and price points, Google is enabling developers and businesses to integrate AI agents into various applications. This move could lead to increased automation, improved productivity, and innovations across industries as AI agents become more capable of handling complex, multi-step tasks.

5 MINUTE AI LEARNING

YOUR 5 MIN AI LEARNING

Search for the Needle in Your Bookmark Haystack with mindECHO

Okay, for this week's AI learning section, we're diving into mindECHO, a seriously cool search tool that's all about streamlining your digital life. If you're like me and have a million bookmarks and random web pages bookmarked across different browsers, mindECHO is about to become your new best friend. It lets you quickly search across all that saved content to find exactly what you need, without spending ages manually sifting through everything, which you and I both know takes way too long. I just want to find my banana bread recipe! 

We could all use a little help getting organised, and mindECHO might just be the tool we've been waiting for. So, the next time you need that recipe you saved years ago for this week’s dinner or a tutorial on how to change your car’s engine oil, give mindECHO a try!

LATEST AI TOOLS

Reap.video 

A tool to convert long videos into short videos optimised for social media clips (Freemium)

“Repurpose your content for Instagram Reels, YouTube Shorts, and LinkedIn posts with automatic resizing for all formats: square, vertical, and horizontal. Refine your videos effortlessly with AI-driven editing tools designed to save you time and elevate your content quality.”

Unbaited 

A browser extension to filter engagement bait from social media feeds (Free)

“Unbaited is a browser extension that helps you filter out engagement bait and inflammatory content from your X (formerly Twitter) feed. It uses AI to analyze tweets in real-time and hides content designed to provoke emotional responses or increase engagement through controversial topics.”

THEO 

A tool to reshape business content into AI-ready formats for enhanced AI assistant responses (Paid)

“Turn your website & docs into one structured 'cheat sheet' and use it to turn you AI into a strategic partner. Perfect for founder's strategic ideation, planning & analysis.”

AI HIGHLIGHTS

High-Stakes AI Summit in Paris

Next week, political figures and AI pioneers will gather in Paris to participate in a crucial summit focused on addressing the growing influence of artificial intelligence and its wide-ranging disruptive potential. The discussions will centre on navigating the complex challenges and opportunities presented by AI, as well as establishing frameworks to ensure its responsible development and deployment across society.

New AI Model Rivals OpenAI with Minimal Resources

A new open-source AI reasoning model called S1 rivals OpenAI's models by using a simple technique of pausing during its reasoning process, demonstrating that significant AI advancements can be achieved with minimal resources. Despite concerns about unauthorised use and the challenges of accuracy, affordable models like S1 could become beneficial for on-device processing and increase the demand for computing resources as AI permeates various aspects of life.

Trump’s Plans for the Education Department 

Donald Trump is considering shrinking or eliminating the Department of Education, with his administration reportedly preparing an executive action and Elon Musk's DOGE using AI to identify potential spending cuts within the department. Trump aims to shift control of schools back to the states, though he cannot legally shut down the department without congressional approval, and the future of the $1.6 trillion (approximately £1.3 trillion) student loan program remains uncertain.

Google Photos Adds AI Watermarks

Google Photos is adding digital SynthID watermarks to images modified with the Magic Editor’s generative AI feature to help users identify altered content. The watermark, which is imperceptible, is integrated directly into images, audio, video, and text, allowing for the identification of AI-created or modified content.

AI IMAGE OF THE WEEK

For this week’s AI image, we here at MegaSuperAI felt inspired by WWE’s Royal Rumble pay-per-view (which was actually shown on Netflix!) So, let’s see how an AI image generator imagines a wrestling match.

The prompt used to generate this image on Microsoft Designer: 

“Hyper-realistic image of two wrestlers in a wrestling ring with one throwing the other over the top rope. The view is from the outside of the ring looking in.”

Send us your AI generated art for a chance to be featured.

FEEDBACK

• Let us know how we did - shoot us a quick email. We read every single one (no AI in sight!).

• If you enjoyed this newsletter please share with an awesome friend or two.

• If you are that awesome friend then hit the below button: