• AI Academy
  • Posts
  • 🤖 Google Shipping AI Features At Record Pace

🤖 Google Shipping AI Features At Record Pace

Plus: Italian AI Newspaper and Corporate AI Disconnects

Hello AI Enthusiast,

This week brings us an interesting mix of progress and challenges in the AI world. Google continues to enhance its Gemini offerings with features that quietly impress (especially image generation), while Italy's Il Foglio claims to be the first newspaper to publish an entirely AI-generated edition. Meanwhile, a new survey reveals the growing disconnect between how executives and employees view AI implementation in their companies. Let's explore these developments and what they mean for the future of AI adoption.

The Big Picture 🔊

Google Expands AI Capabilities

Google has expanded Gemini 2.0 Flash native image generation to developers. The technology excels at conversational image editing and accurately rendering text in images. Meanwhile, Gemini received significant upgrades: a 1M token context window for Advanced users, widely available Deep Research, integration with more Google apps (Calendar, Notes, Tasks, soon Photos), and free Gems for all users. Additionally, Gemini now features Canvas for real-time collaboration on documents and code, and Audio Overview, which transforms documents into engaging, podcast-style AI discussions for easier learning on the go.

Gianluca Belloni
Gianluca BelloniCMO and Marketing Nomad

Gemini's text rendering in images impressed us, unlike most AI systems. Yet we're puzzled by Google highlighting watermark removal, which raises obvious copyright concerns.

Despite releasing powerful features at competitive prices and having deep integration with its widely-used services, Google struggles to generate the same excitement as competitors. There’s a missing piece in Google’s AI marketing strategy.

Il Foglio Publishes World's First AI-Generated Newspaper Edition

Italy's conservative daily Il Foglio has released what it claims is the world's first entirely AI-generated newspaper edition. The four-page "Il Foglio AI" supplement features AI-written articles, headlines, quotes, and summaries, with journalists only "asking questions and reading answers." Content included stories about Trump, Putin, the Italian economy, and relationship trends among young Europeans. The articles lacked direct human quotes.

James Varnham
James VarnhamCEO and Rainmaker

This AI newspaper raises key questions about journalism's future. Did AI choose the topics or only write them? News should be facts, not personalized content, and AI-generated articles without human quotes further blur this line.

Also, many readers might not even notice they're reading AI-generated content, making our information landscape even more confusing.

AI Creating Deep Divide Between Executives and Workers

2025 Writer survey reveals a stark disconnect in AI adoption perceptions. While 89% of executives claim their company has an AI strategy, only 57% of employees agree. Half of executives believe AI is "tearing their company apart," with 59% seeking jobs elsewhere. Worker resistance stems from replacement fears and poor tools - 41% of younger employees admit to sabotaging AI initiatives, and 35% pay out-of-pocket for better AI tools.

Helin Yontar
Helin YontarCPO and Polyglot

This survey confirms what we've seen firsthand, companies are failing at AI communication. We regularly see employees using ChatGPT despite company-purchased alternatives because companies rarely justify their tool selections.

Without understanding "why" behind decisions, younger workers especially resist adoption, quietly using their preferred tools instead.

Is your company facing this AI divide? Our training programs bridge the gap by helping teams understand not just how to use AI tools, but why they matter. Whether you use Copilot, Gemini, Claude, or ChatGPT, our trainings align leadership vision with practical implementation. Don't waste your AI investment, learn how we can help your organization embrace AI effectively.

Bits and Bobs 🗞️

  • Google announced the development of "open" AI models for drug discovery called TxGemma, aiming to streamline the drug development process.

  • Elon Musk’s AI company, xAI, has acquired the startup Hotshot, which specializes in AI-powered video generation models, possibly indicating xAI’s intention to develop its own video creation tools.

  • Mistral Small 3.1 is the first open-source AI model to surpass proprietary small models in handling text, multimodal inputs, and multiple languages efficiently, and it is available for download under an Apache 2.0 license.

  • Mirage by Captions is the world's first AI model designed to generate complete videos from just a prompt, creating original actors and content without licensing restrictions, and simplifying video ad production for marketers.

  • Amazon Web Services (AWS) has offered one of its long-term cloud customers access to advanced chips intended to deliver enhanced computing power for AI applications.

  • Nvidia CEO announced the Blackwell Ultra chip family, expected in the second half of 2025, designed to improve AI inference and reasoning capabilities.

  • OpenAI’s o1 and o3-mini models now support Python-powered data analysis in ChatGPT, handling tasks such as performing regression analysis on test data, creating visual representations of complex business metrics, and running scenario-based simulations.

That's a wrap on our newsletter! Before you go, here’s a quick recap of our offerings:

Catch you next week! 👋