Apple just dropped a wave of AI-powered features across its ecosystem at WWDC 2025—ushering in a new era of on-device intelligence, smarter automation, and seamless translation. Here’s what stood out:
🈂️ Live AI Translation Comes to iOS, iPad, and Watch
iOS 26 adds real-time AI translation across Messages, Phone, and FaceTime—even with non-Apple users. Translations appear as captions or spoken aloud during calls, and messages auto-translate while typing. Also coming to iPads and Apple Watches. Supported languages TBD.
🚀 Shortcuts Gets an AI-Powered Upgrade
The Shortcuts app now supports on-device LLMs via Apple Intelligence, letting users automate tasks with AI-generated text, image creation, and more. It’s a major leap toward making Siri more useful—and automation more approachable.
👨💻 Apple Opens Foundation Models to Developers
Apple’s new on-device AI framework allows devs to use its models for offline, privacy-first features. Built for Swift, it’s already being tested by apps like Day One and AllTrails. A public beta is coming next month.
⌚ Smarter, Sleeker Apple Watch with watchOS 26
Apple’s watchOS 26 brings “Liquid Glass” UI, an AI-powered fitness coach (Workout Buddy), enhanced Smart Stack widgets, live translation, music suggestions, and a new gesture to manage notifications—making the Watch more helpful and intuitive than ever.
Enjoying our newsletter? Forward it to a colleague—
it’s one of the best ways to support us.
☝️ POWERED BY NEUBIRD
Incident Alerts Eating Into SLAs? Go from Alert to Remediation in Minutes with Your AI SRE
Every minute counts during an incident. Hawkeye, your always-on AI SRE agent, diagnoses issues instantly, reduces MTTR (mean time to resolve) by up to 90%, and frees engineers from the grind of endless firefighting. Built for enterprise IT, backed by Mayfield and Microsoft’s M12. Now available on the Datadog Marketplace.
Your always-on AI teammate is now available for hire on the Datadog Marketplace. No dashboards. No prompts. Just real-time resolution. Learn more here!
👾 FORWARD FUTURE ORIGINAL
A2A vs. MCP
Today’s smartest AI models still have one big flaw: they work alone.
That’s changing fast.
Google’s Agent-to-Agent (A2A) and Anthropic’s Model Context Protocol (MCP) are two emerging frameworks that turn solo models into connected systems—able to coordinate, delegate, and tap into live data and tools.
This article breaks down how they work, how they differ, and why the next wave of AI will be less about chatbots—and more about digital ecosystems.
→ Read the full explainer on A2A vs. MCP.
🏢 ENTERPRISE
AI Blurs the Office Line as SAP and Salesforce Vie for Enterprise Supremacy
SAP and Salesforce, long dominant in back-office and front-office enterprise software, are now squaring off as AI redraws the boundaries between their domains. With AI agents that analyze and act on data across business functions, both companies aim to become the go-to interface for enterprise productivity. SAP's broader operational reach and tighter grip on customer data give it a current edge, with investors rewarding its integrated strategy.
Meanwhile, Salesforce faces slowing growth, share price declines, and skepticism over its $8B Informatica acquisition. As both firms race to blend CRM and ERP into a seamless AI experience, the old lines between front and back office are vanishing fast. → Read the full article here.
🖥️ HARDWARE
AMD Buys Brium to Loosen NVIDIA’s Grip on AI Hardware Ecosystem
AMD has acquired stealth-mode startup Brium, a company focused on optimizing AI inference across diverse hardware platforms—a direct challenge to NVIDIA’s entrenched dominance. While most AI software is still tuned for NVIDIA GPUs, Brium’s tools aim to make it easier for developers to run models on alternatives like AMD’s Instinct chips.
The acquisition aligns with AMD’s broader strategy to create a high-performance, open AI ecosystem and reduce reliance on a single vendor. It’s AMD’s fourth AI-centric deal in two years, reinforcing its methodical approach to expanding influence in the AI hardware space and making room for real competition in inference workloads. → Read the full article here.
⚛️ SCIENCE
Small Quantum Processor Outperforms Classical Algorithms in Machine Learning
Researchers from the University of Vienna have shown that small-scale photonic quantum processors can already outperform classical computers on certain machine learning tasks. In a study published in Nature Photonics, the team used a quantum photonic circuit to run a data classification algorithm, demonstrating fewer errors than its classical equivalent.
The results suggest that quantum advantage doesn’t require futuristic hardware—today’s limited systems can still deliver meaningful performance gains. Photonic platforms also offer potential energy savings, a growing concern as machine learning workloads become more power-hungry. The work opens new paths for both quantum computing and more efficient classical algorithm design. → Read the full paper here.
🛰️ NEWS
What Else is Happening
🚫 China Blocks AI During Exams: AI apps like Qwen, Doubao, and Kimi disabled image recognition to curb cheating as 13M students sit for China’s grueling gaokao college entrance exams this week.
🏗️ Amazon Bets Big on PA Data Centers: Amazon will invest $20B to build AI and cloud campuses in Pennsylvania—its largest U.S. investment yet—creating 1,250+ jobs near Philly and Luzerne County.
©️ 130K Scripts Trained AI Models: A BFI report reveals film & TV scripts were used without consent to train AI, raising alarm over copyright, creator pay, and the future of screen industry economics.
🦉 Duolingo Faces Backlash Over AI Pivot: CEO Luis von Ahn’s “AI-first” shift sparked user outrage, subscription drops, and PR damage—forcing a public walk-back and damage control campaign.
🏅 Microsoft Adds AI Safety Scores: To boost trust, Microsoft will rank AI models on safety—flagging risks like hate speech and misuse—alongside cost and quality for Azure cloud customers.
📟 PROMPT OF THE WEEK
SWOT Analysis Assistant
You are a business analysis assistant. The user will provide a brief description of their business or project. Generate a SWOT analysis in four parts: Strengths, Weaknesses, Opportunities, Threats.
Under each part, list 3-5 bullet points that apply to the given context. Strengths/Weaknesses should be internal factors (e.g., resources, expertise, challenges), and Opportunities/Threats external factors (e.g., market trends, competition).
Keep the analysis concise and factual in tone.📽️ VIDEO
Google CEO Sundar Pichai on Gemini, Self-improving AI, and World Models
In case you missed it, Matt sat down with Google CEO Sundar Pichai to explore Gemini’s evolution—a window into the future of intelligent systems. Get the full scoop! 👇
🧰 TOOLBOX
Smarter Workflows, Sharper Insights, and Streamlined Collaboration
🎯 Flux Kontext: Boost team efficiency with context-aware tasks, real-time insights, and seamless system integration for smarter workflows.
📰 Every: Dive into expert-led newsletters on business, productivity, and strategy—actionable insights from top operators and creators.
🛠️ Basecamp: All-in-one tool for managing projects, tasks, and team chats—streamline collaboration and stay organized with ease.
👉️ Find more AI tools: Browse the Forward Future AI Tool Library
That’s a Wrap!
📢 Want to advertise with Forward Future? Reach 500K+ AI enthusiasts, tech leaders, and decision-makers. Let’s talk—just reply to this email.
🛰️ Want more Forward Future? Follow us on X for quick updates, subscribe to our YouTube for deep dives, or add us to your RSS Feed for seamless reading.
Thanks for reading today’s newsletter—see you next time!
Matthew Berman & The Forward Future Team
🧑🚀 🧑🚀 🧑🚀 🧑🚀
