AI for Business 2026: MIT Wireless Breakthrough, InPost Shopping Assistant, Perplexity Premium Sources, and Hollywood AI Trends
Dear Friend,
Last week, in Applied AI we spent time inside presentations and honestly, it was one of those “this saves hours” sessions. The big takeaway? AI tools like Canva, Gamma, and Google Slides are getting faster and more practical, and they still need direction. The best results came when we brought a clear outline first, then let AI build from it.
Let’s get into the stories worth your time this week…
MIT’s AI Breakthrough Lets Systems “See” Through Walls
MIT researchers developed a generative AI-powered wireless system that can detect objects and people through obstructions like walls. This advancement improves accuracy significantly compared to older sensing technologies.
My take: This is one of those innovations that feels futuristic but has very real operational impact. Businesses that rely on physical environments like retail, manufacturing, or healthcare should be paying attention now, not later. Think about how you could reduce your cost-to-serve with this down the line. This opens the door for smarter logistics, security systems, and even healthcare monitoring without invasive hardware. Think warehouses tracking inventory in real time or safer elder care solutions at home.
Creepy or useful?
Read more
InPost Builds AI Shopping Assistant to Compete with Marketplaces
InPost is developing an AI-powered shopping assistant aimed at improving customer experience and competing with large e-commerce platforms. The tool helps users discover products more efficiently and personalize their shopping journey.
My take: If you’re in e-commerce and not thinking about AI-assisted discovery, you’re already behind. The real opportunity here is reducing friction in the buying process while increasing average order value and conversion rates. If your store is easy for agents to discover and purchase from even smaller sellers can now compete by offering smarter buying experiences.
Let us know if you need help with your AI SEO / GEO, [email protected]
Read more
Hollywood Experiments with Generative AI Featuring Val Kilmer
A new film project uses generative AI to recreate Val Kilmer’s voice and presence, showing how AI is reshaping content production in entertainment. Should they be doing this?
My take: This is where AI meets intellectual property and regulation. Businesses should be excited about the creative possibilities but also cautious. Ensure your outputs are defensible. This signals major changes in content creation, marketing, and licensing. Brands can explore scalable content production while also navigating new legal and ethical considerations. Presumably Val’s estate gave consent and will be receiveing royalities.
AI Tackles March Madness Predictions with Data-Driven Picks
AI models are being used to predict NCAA tournament outcomes, analyzing massive datasets to improve bracket accuracy. Last year AI did a great job here, we’ll see if there’s a repeat.
My take: Sports are often the testing ground for broader AI adoption. If a model can handle the chaos of March Madness, imagine what it can do with your business data to predict churn or lifetime value. This is a great example of predictive analytics in action. The same approach can be applied to sales forecasting, demand planning, and risk assessment.
Read more
All right, let’s talk about the future of your team. By 2027, every employee will need to be “AI smart.” This is someone who can apply artificial intelligence within their specific role to enhance traditional work patterns. What skills will every employee need by 2027?
Gartner predicts that 80% of the workforce will require upskilling by 2027 due to the rise of AI-native software and agentic systems. At AI Smart Ventures, we’ve documented that employees with high AI literacy achieve 50% average time savings by moving from “blank page” starts to AI-assisted refinement.
The goal of upskilling is not to turn every employee into a data scientist. It is about building a future-proof workforce that knows how to steer AI agents toward the right context and constraints. The shift from “doing the work” to “directing the work” is the core transition for the next two years.
Here are the 5 essential skills your team needs:
1. Prompt Engineering 2.0: Advanced AI Communication
Prompt engineering is moving beyond simple requests. Employees must learn how to guide AI platforms with precision-providing context, examples, and output constraints. This skill directly impacts work by allowing faster iteration on workflows.
2. Critical Evaluation: AI Output Verification
You have to verify the data. Sometimes the information just doesn’t exist online. Employees must be able to judge the quality, accuracy, and safety of AI-generated work to ensure your firm maintains its reputation. Lack of output verification is one of the top reasons AI initiatives fail to scale.
3. Data Literacy: Understanding AI-Generated Insights
This involves understanding how to interpret dashboards and translate complex data patterns into actionable business strategy. Professional services firms report that employees with strong data literacy identify revenue opportunities 40% faster.
4. Automation Workflow Design: Process Reimagination
This is the skill of outlining existing procedures and rethinking how work is completed with AI assistance. Map current processes, identify repetitive decision points, and design intelligent agents that handle routine tasks autonomously.
5. Ethical Judgment and AI Governance
Ethical judgment involves navigating the risks of data privacy, client trust, and global AI regulations. You need clear guardrails. Organizations must understand what data can and cannot be shared with AI systems to avoid “Shadow AI.”
Training: Build vs. Buy
When deciding how to develop these skills, custom training programs deliver significantly higher retention rates (70-85%) because they address firm-specific workflows rather than generic AI concepts. Organizations maximizing tools they already own, like Google Workspace or Microsoft 365, spend 40 to 60% less on training while achieving better adoption outcomes.
Read the full guide on our blog here.
AI Tools to Explore This Week
- Jasper.ai: Great for teams creating consistent, high-quality marketing content while keeping brand voice aligned.
- HeyGen: Create professional videos with AI avatars and voiceovers. Perfect for ads, training, or social media content without a full production team.
- Read.ai: Helps track meetings, emails, and messages with AI-generated summaries so nothing falls through the cracks.
How to Create Your Own Songs in SUNO
For some inspo, here’s a little playlist.
Step 1: Create an Account
- Go to the website: Open your browser and navigate to suno.com.
- Sign up: Click “Make a song” or “Sign In” at the bottom left. You can sign up using your Google, Microsoft, or Discord account.
- Check your credits: Free accounts typically receive 50 daily credits, which is enough to generate 10 songs (Suno generates two versions per prompt, costing 10 credits total).
Step 2: Go to the “Create” Tab
- Look at the left-hand menu and click on the Create button. This opens the control panel where you will prompt the AI.
Step 3: Choose Your Mode
You have two main ways to create a song in Suno: Basic Mode and Custom Mode.
Option A: Basic Mode (Quick & Easy)
- Leave the “Custom” toggle switch off.
- Song Description: Type a description of the song you want. For example: “An upbeat synth-pop song about drinking coffee on a rainy Monday morning.”
- Instrumental: Toggle this switch on if you only want music without vocals.
Option B: Custom Mode (More Control)
- Toggle the Custom switch to the on position.
- Lyrics: You can type or paste your own lyrics here. If you are stuck, you can type a topic and click “Make Random Lyrics” to have the AI write them for you. Use tags like
[Verse],[Chorus], and[Guitar Solo]to guide the structure. - Style of Music: Enter the genres, vibes, and instruments you want (e.g., acoustic indie folk, male vocal, melancholic).
- Title: Give your song a name.
Step 4: Generate the Song
- Once you are happy with your prompt, click the Create button.
- Wait a few moments. Suno will generate two different versions of your song based on your instructions.
Step 5: Listen, Edit, and Share
- Listen: Click the play button on the newly generated tracks in your library on the right side of the screen.
- Extend: If you like a song but it cuts off too early or you want to add more verses, click the three dots (
...) next to the song and select Extend. This lets you add more lyrics or music from the point where the song ended. - Download: Click the three dots, select Download, and choose whether you want an Audio file (MP3/WAV) or a Video file (which includes the song’s cover art and scrolling lyrics).
Have fun!
Join the Conversation
Got a burning question, a fresh take, or just want to share your latest AI wins? Hit us up at [email protected]. Your insights keep this community growing and thriving!
See you in the Lab,
-Nicole A. Donnelly
Founder, AI Smart Ventures
AI Strategy, Consulting, Training, and Implementation
