Roberto Munoz

Fractional Chief Comms Officer | Strategic Communications Leader | President, Munoz Comms

I'm CEO & Founder of Munoz Comms, a strategic communications consultancy, with 20+ years' experience protecting and promoting top brands--from startups to Fortune 500s.

HSBC • Tesco • Fannie Mae • Royal Bank of Canada • ABB • Bristol Myers Squibb • Axiom legal • Jaguar Land Rover • Motorola

Munoz Comms is designed to help brands navigate growth, change and transformation. Developing strategic communications that deliver measurable results. 

Our core services: 
→ Comms Strategy
→ Thought Leadership
→ Narrative & Positioning
→ Change Communications
→ Fractional Chief Comms Officer

My track record: 
• Global C-Suite executive who can navigate companies (multinational and start-up). 
• Industry leader who deeply understands the ever-changing media landscape.
• Trusted C-Suite leader and CEO confidant. 
• Transformational change agent with decades of in-house experience. 

What I do best: 
- Tell stories. 
- Deliver results.
- Focus on outcomes.
- Solve tough problems.
- Develop talented teams. 
- Shift stakeholder perceptions.  

Let’s connect: 
roberto@munozcommunications.com
www.munozcommunications.com

Mixing Board Member

Mixing Board is an invite-only community of senior marketing and communications leaders who share experience, expertise, and mentorship to elevate our practice.

Join 500k+ AI Tool Report subscribers and start learning AI in 5 minutes a day.

AI is not just the future. It’s your competitive edge—today.

I help ambitious professionals and companies stay informed, and ahead with AI. 

As founder of The AI Report, I lead the #1 B2B AI newsletter helping 400,000+ readers—from NASA, Tesla, Meta, and more—master the real-world use of AI in 5 minutes a day.

What I focus on as CEO:

📈 Driving company vision, growth strategy, and business model innovation
🧠 Building world-class editorial and research systems across content, product, and training
🤝 Leading partnerships with global brands and AI providers
👥 Hiring and enabling exceptional operators to scale execution

AI isn’t a trend. It’s the most important lever in your business—and most are still pulling it wrong.

If you're ready to stop dabbling and start scaling, I can help.

Let’s talk.

Martin Crowley

AI hallucinates facts constantly.
Here's how to catch it before it matters.

1. Critically Evaluate Every Output

AI doesn't think or reason.
It predicts words based on patterns.

What this means:
→ No inherent truth verification
→ No ability to distinguish fact from fiction
→ No understanding of what it's saying

Your job:
Apply human judgment to every output.
Treat AI like a smart intern who sometimes makes things up.

2. Diversify Your Sources

Never trust AI alone on critical facts.

Cross-check with:
→ Industry experts
→ Peer-reviewed publications
→ Primary sources
→ Multiple AI tools

One source = potential hallucination
Three sources = likely accurate

3. Use Retrieval-Based Tools

Some AI tools use RAG (Retrieval-Augmented Generation).

What it does:
→ Pulls from trusted documents first
→ Generates answers based on those sources
→ Cites where information came from

Examples:
→ ChatGPT with uploaded documents
→ Claude Projects with files
→ Perplexity with source citations

Research shows RAG improves accuracy and trust.

4. Use Clear and Structured Prompts

Vague prompts = vague (often wrong) answers.

Bad prompt:
"Tell me about the market."

Good prompt:
"Analyze Q4 2024 SaaS market trends. Explain your reasoning step-by-step. Cite specific data points."

The technique:
Chain-of-Thought Prompting forces AI to show its work.
Exposes logical gaps and unsupported claims.

**5. Adjust Temperature Settings**

Temperature controls randomness.

Low temperature (0 to 0.3):
→ Focused and factual
→ Consistent outputs
→ Best for: Data analysis, reports, technical content

High temperature (0.7 to 1.0):
→ Creative and varied
→ More imaginative
→ Best for: Brainstorming, storytelling, creative work

For critical accuracy, always use low temperature.

The pattern that works:

Step 1: Use structured prompts with low temperature
Step 2: Request step-by-step reasoning
Step 3: Cross-check with trusted sources
Step 4: Apply human judgment to final output

Most people skip steps 3 and 4.
That's when expensive mistakes happen.

AI is a tool, not a source of truth.
Treat it accordingly.

P.S. Want to learn more about AI?

1. Scroll to the top
2. Click "Visit my website"
3. Sign-up for our free newsletter

You don't need to be technical. Just informed.

 5 Ways to Stop AI From Lying to You

Most people use ChatGPT for work.

Smart people use it for self-awareness.

You pay $200/hour for a therapist.
They help you understand your patterns.
Process your emotions.
Navigate workplace dynamics.

What if you uploaded your personality to an AI?

ChatGPT can become your Personal OS with the right context.

The 4-step system that turns AI into your emotional intelligence coach:

1. Gather your personality framework data

Collect your Enneagram results.
Myers-Briggs assessment.
DISC profile.
Human Design chart.

Any framework that explains how you think.

2. Create a dedicated ChatGPT Project

Upload all your personality documents.
This becomes your "Personal OS."

The AI now understands your default patterns.
Your triggers. Your communication style.

3. Analyze workplace relationships

Upload your boss's operating manual.
Or a colleague's personality profile.

Ask: "How should I work with Morgan considering he's an Enneagram 5 and I'm a 7?"

The AI maps the interaction dynamics.
Shows you where friction will happen.
Suggests strategies that work for both types.

4. Use it for real-time emotional coaching

Something pisses you off at work.

Tell your AI: "I'm really mad someone's trying to control my project."

It validates your feelings.
Then reframes the situation.
Suggests responses aligned with your personality type.

Most self-help is one-size-fits-all.
"Be more assertive" works for some people.
Backfires spectacularly for others.

Your Personal OS knows YOUR patterns.
It gives you strategies that match your wiring.

You're paying therapists to learn your personality from scratch every session.
When AI can load your entire psychological profile in 30 seconds.

Found this helpful? Follow Martin Crowley.

How to Build Your Own AI Therapist in 10 Minutes

Most people still use ChatGPT alone.
That model just became obsolete.

ChatGPT now runs apps inside conversations.

Here's what just happened:

You're chatting about weekend plans.
ChatGPT suggests Spotify mid-conversation.
Builds your party playlist without leaving.

No tab switching.
No copy-pasting.
Just conversation → action → done.

Real examples that work today:

"Find me a hotel in Paris"
→ [Booking.com](http://booking.com/) appears with live results
→ Filter by price inside the chat
→ Book without opening browser

"I'm buying a house in Austin"
→ Zillow loads interactive map
→ Shows listings in your budget
→ Schedule tours through chat

"Turn this outline into slides"
→ Canva builds full presentation
→ Edit design through conversation
→ Export when ready

The apps available now:

→ [Booking.com](http://booking.com/) (travel planning)
→ Canva (instant slide decks)
→ Coursera (learning while chatting)
→ Expedia (flight + hotel booking)
→ Figma (design collaboration)
→ Spotify (playlist generation)
→ Zillow (house hunting)

Coming soon:

→ AllTrails for hiking routes
→ OpenTable for reservations
→ Peloton for workouts
→ Target for shopping
→ Uber for rides

Discovery just changed forever.

You don't browse app stores anymore.
Apps appear when context demands them.

Talking about dinner plans?
→ OpenTable surfaces automatically

Discussing workout goals?
→ Peloton appears with programs

Planning a trip?
→ Expedia + Booking stack together

This isn't an app store.
It's conversation-native computing.

The platform war just restarted.

P.S. Want to learn more about AI?

1. Scroll to the top
2. Click "Visit my website"
3. Sign-up for our free newsletter

ChatGPT Now Has an App Store

You're getting bad AI outputs because your prompts are lazy.

Here's how to fix it:

Rule 1: Provide Context

Bad prompt:
"What's the best time for New England fall foliage?"

Good prompt:
"You are an experienced wildlife biologist specializing in trees. Based on recent USA weather patterns, predict the best fall foliage season for New England and explain it to kindergarteners."

The difference:
→ Role sets expertise level
→ Parameters focus the answer
→ Audience shapes the language

Rule 2: Be Specific

Bad prompt:
"Tell me about climate change."

Good prompt:
"Discuss the economic implications of climate change in developing countries over the next decade."

The difference:
→ Year specified
→ Region defined
→ Aspect clarified
→ Scope limited

Specificity = quality.
Vague inputs = vague outputs.

Rule 3: Build On the Conversation

AI remembers your chat history.
Use it.

First prompt:
"Explain quantum entanglement to kindergarteners."

Follow-up:
"Make it funnier."

Another follow-up:
"Now explain it to college English majors using analogies they'll understand."

No need to repeat context.
Just keep building.

Real example:

Prompt 1:
"You are a sales coach. Write a cold email to CFOs about budget software."

Prompt 2:
"Make it shorter, under 50 words."

Prompt 3:
"Add a case study from the finance industry."

Prompt 4:
"Rewrite for VP of Finance instead."

Each builds on the last.
No starting over.

Why this works

Context = relevant answers
Specificity = accurate answers
Building = refined answers

Most people ask once and quit.
Experts iterate until perfect.

The formula:

Step 1: Set role and context
Step 2: Add specific constraints
Step 3: Iterate until satisfied

Try this on your next prompt.
You won't recognize the difference.

P.S. Want to learn more about AI?

1. Scroll to the top
2. Click "Visit my website"
3. Sign-up for our free newsletter

3 Prompt Engineering Rules That 10x Your AI Results

Healthcare doesn’t need “smarter” AI.

It needs less friction.

Anthropic’s Claude for Healthcare (and OpenAI’s ChatGPT Health days earlier) signal the same shift:

AI is being formalized where people were already using it, under pressure, without structure.

The real opportunity isn’t medical advice.

It’s administrative decisions repeated millions of times:

 • prior authorizations

 • documentation

 •coverage checks

 • reporting

That’s where time, money, and clinician energy are leaking.

But this race isn’t just about speed.

In healthcare, hallucinations become liability, not bugs.

So the advantage won’t come from model performance alone.

It will come from:

 • verified data access

 • auditability

 • human-in-the-loop controls

 • trust earned over time

AI is entering healthcare the same way it entered finance:

slowly, carefully, and where the ROI is obvious.

💬 Where should humans remain accountable even if AI becomes more accurate?

Anthropic Launches Claude For Health Care

Your inbox is quietly becoming a task manager.

Google has unveiled an AI-powered inbox for Gmail that replaces the traditional email list with summaries and suggested actions. Instead of sorting by recency, Gmail now prioritizes what it thinks you should do next.

For operators and leaders, this is a real workflow shift.

If it works, email triage becomes faster and more decision-focused.

If it fails, you risk outsourcing attention and judgment to an algorithm.

→ Email is no longer just communication. It’s becoming an AI-led priority queue.

#Gmail #GoogleAI #EmailProductivity #AIWorkflows #FutureOfWork

This isn’t about AI replacing artists. It’s about redefining discovery.

Universal Music Group partnering with NVIDIA on Responsible AI is a clear signal: the next phase of music tech is creative-first, not automation-first.

By combining NVIDIA’s Music Flamingo with UMG’s deep catalog, the goal goes beyond personalization, toward understanding music at the level of harmony, structure, lyrics, and cultural context.

The real test won’t be the announcement.

It’ll be whether attribution, protection, and quality hold up at scale.

→ Execution will matter more than intent.

#MusicTech #ResponsibleAI #CreativeEconomy #ProductStrategy #TheAIReport

OpenAI is formally entering healthcare with ChatGPT Health.

OpenAI says more than 230 million people ask health-related questions in ChatGPT every week, and now those interactions are being structured into a dedicated ChatGPT Health experience.

The focus isn’t diagnosis or treatment. It’s preparation:
helping users understand test results, track trends from wearables, and walk into medical appointments with better questions.

The most important feature isn’t the model, it’s the trust layer:
-Separate health space
-Compartmentalized memory
-Added encryption

Health conversations not used to train foundation models

→ This will be valuable if privacy protections hold at scale.

#OpenAI #ChatGPTHealth #AIinHealthcare #DigitalHealth #HealthTech

Most people write vague prompts and wonder why AI fails.
Here's how to get professional results every time.

1. Clean Skin Without Losing Texture

Bad prompt: "Make skin look better"

Good prompt:
"Reduce small blemishes and even tone while keeping all natural pores and texture visible. Smooth only problem areas under 3mm. Preserve eye sharpness and face structure completely."

2. Studio-Quality Portrait Lighting

Bad prompt: "Fix the lighting"

Good prompt:
"Add soft directional light from upper left at 45 degrees. Lift shadow detail by 15%, add gentle highlight on cheekbones, maintain original skin color accuracy."

3. Film-Style Color Grading

Bad prompt: "Make it cinematic"

Good prompt:
"Apply teal-orange color split with cooler shadows and warmer highlights. Keep skin tones neutral. Add subtle grain at 8% opacity. Increase midtone separation without crushing blacks."

4. Remove Background Distractions

Bad prompt: "Clean up the background"

Good prompt:
"Remove objects and people in background only. Maintain original blur and depth of field. Fill removed areas with matching texture and lighting. Keep all subject edges sharp."

5. Fix Small Imperfections Only

Bad prompt: "Remove blemishes"

Good prompt:
"Target spots, stray hairs, and dust marks smaller than 4mm only. Use texture-matched healing. Leave skin texture, freckles, and natural features completely unchanged."

6. Enhance Makeup Subtly

Bad prompt: "Make makeup pop"

Good prompt:
"Deepen existing eye makeup by 12%, sharpen liner edges slightly, increase lip saturation by 8%. Avoid adding shine. Keep skin texture and pores fully visible."

7. Correct Outdoor Color Cast

Bad prompt: "Fix outdoor colors"

Good prompt:
"Remove green color cast from reflected foliage. Warm midtones by 6%. Recover blown highlight detail in sky. Keep gradient smooth with no banding. Preserve tree and grass detail."

8. Black and White Conversion

Bad prompt: "Make it black and white"

Good prompt:
"Convert to monochrome with strong contrast and deep blacks. Keep shadow detail visible. Boost texture clarity. Add fine grain at 10%. Place skin tones in upper midrange for luminance."

9. Product Photo Enhancement

Bad prompt: "Polish this product"

Good prompt:
"Remove dust and surface marks only. Sharpen product edges selectively. Enhance material texture without changing color. Keep color within 3% of original for brand accuracy."

10. Social Media Optimization

Bad prompt: "Make it Instagram ready"

Good prompt:
"Optimize for mobile viewing by increasing sharpness perception by 10%, boosting saturation 7%, cropping to 4:5 ratio. Preserve face detail. Ensure colors display consistently across devices."

P.S. Want to learn more about AI?

1. Scroll to the top
2. Click "Visit my website"
3. Sign-up for our free newsletter

 10 Gemini AI Photo Editing Prompts 

Notepin S isn’t trying to be a better notebook. It’s trying to own the meeting.

By pairing a wearable recorder with a desktop client, Notepin S is positioning itself as a full meeting workflow, not just a transcription tool.

The differentiator isn’t the hardware alone.

It’s the combination of physical capture, digital meeting ingestion, and structured outputs that turn conversations into usable notes.

That places it alongside tools like Granola, Otter, and Fireflies but with a tighter loop between what’s said and what gets saved.

→ In meeting software, capture without context doesn’t scale.

#MeetingProductivity #WorkplaceTechnology #WorkflowDesign #AIAtWork #TheAIReport

taplio

The Battle

Roberto Munoz

Martin Crowley

Followers

Posts

Engagements

Likes

Performance ScoreMartin Crowley's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Roberto Munoz

Martin Crowley

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Martin Crowley's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreMartin Crowley's LinkedIn Performance