Using AI to Tighten the Interpretation Loop at Work

As a business generalist, one of the most common use cases for applying technology in my own work is to help me interpret, synthesize, and communicate with people who have vastly different communication styles. This skill has become particularly important as a fractional worker, since I dip in and out of various project teams, industries, and domains of expertise.

On these teams, I often take on the role of a "translator," bridging communication gaps between different groups or even across industries. This iterative process of continuously refining communication by translating ideas and concepts back and forth is what I call the interpretation loop—passing information from one mode or medium to another, then translating it for someone else.

I've naturally taken on this interpreter role in most jobs I've held over the years, but AI has been a game-changer for tightening these interpretation loops, enabling me to connect with people at their level in ways that weren’t possible before.

In this post, I'll share a few examples of how I use AI to bridge these gaps. But first, let’s look at a non-AI example of how the interpretation loop works in everyday communication: Texting your friends.

You can send a lot of content types by text. I have one friend who prefers to text me using short videos, images, or voice memos. This is not my preferred mode of communication. So when he texts me a video or a voice memo, I respond in paragraphs. Despite using different formats, we understand each other and keep the conversation going. Each of us chooses our preferred medium, and this strengthens the interpretation loop between our different communication styles.

In a sense what we are doing is using technology to translate and interpret each other's meaning. We no longer need to use the same platform type to respond. We can speak in whatever medium feels most natural to us.

Layer in AI and it's a whole different ball game. Thanks to natural language interfaces like ChatGPT, if you can tell a story with words, you can generate an image. Conversely, if you are more of a visual thinker, you can share an image with AI, and then generate words to support your creative vision. I've been using this particular interpretation loop a lot in my blog.

Now let's look at how AI might help with a different type of interpretation loop: Creating design concepts on a cross-functional creative team.

If you've got an idea—say, you want to create and conceptualize a new creative, collaborative workspace in Manhattan—it's important to make sure everyone on your team really understands the vision. For instance, in yesterday's blog post, I really wanted to be sure that what I was imagining in my head was the same thing as what people would imagine when reading.

But I'm a writer, not a designer. I think in words, not images. So, I used my words as a “starter prompt” for AI to turn my ideas into visuals. Here’s how that worked:

Starter Prompt: Plain Language
Generate an image of a botanical garden intertwined with an art gallery, offering both nature and creativity. Visitors can stroll through lush, themed gardens while exploring sculptures, photography, and installations along the paths. In the central courtyard, a cafe with outdoor seating invites people to gather, collaborate, or reflect. Workshops on horticulture, painting, and sculpture are held regularly, creating a dynamic intersection of art and nature.

Output from AI: Visual Image
(see below)

The inverse can also be true. If I were a designer who sketched all day but struggled to put ideas into words, my "starter prompt" to AI could be an image. The AI would then translate that visual concept into words for me to share with the team.

Here's what happened when I "reverse engineered" that AI-generated design concept, starting with an image instead of words.

Starter Prompt: Image + Short Text Request
Describe the uploaded image in one paragraph
[insert image above]

Output from AI: Written Text
The scene portrays a vibrant indoor garden space that blends nature with art, creating a tranquil yet dynamic environment. The area is teeming with lush greenery, including trees, ferns, and flowering plants, all intricately arranged around multiple levels connected by winding pathways and stairs. Art installations, such as statues and large abstract paintings, are thoughtfully placed throughout the space, adding layers of creativity to the natural surroundings. People leisurely explore the area, engaging with the art or simply enjoying the serene atmosphere. A central water feature draws attention, while a communal table offers a place for quiet reflection or conversation, seamlessly integrating nature, art, and human interaction. The combination of organic and artistic elements suggests a place of cultural exchange and contemplation.

Whether you start with an image or words, you can easily translate your preferred communication style into one that resonates better with others. This flexibility has many powerful applications.

For instance, let's imagine someone on my team is a native Spanish speaker. Rather than share only the English text, I can also instantly translate it to Spanish. I'm not a Spanish speaker, so I can't vouch for the quality of the translation. Even if it’s not perfect, it likely brings the interpretation loop between us much closer than it would otherwise be.

Starter Prompt: Written Text (English)

Translate the following into Spanish:

The scene portrays a vibrant indoor garden space that blends nature with art, creating a tranquil yet dynamic environment. The area is teeming with lush greenery, including trees, ferns, and flowering plants, all intricately arranged around multiple levels connected by winding pathways and stairs. Art installations, such as statues and large abstract paintings, are thoughtfully placed throughout the space, adding layers of creativity to the natural surroundings. People leisurely explore the area, engaging with the art or simply enjoying the serene atmosphere. A central water feature draws attention, while a communal table offers a place for quiet reflection or conversation, seamlessly integrating nature, art, and human interaction. The combination of organic and artistic elements suggests a place of cultural exchange and contemplation.

Output from AI: Written Text (Spanish)
La escena retrata un vibrante espacio de jardín interior que combina la naturaleza con el arte, creando un ambiente tranquilo pero dinámico. El área está llena de vegetación exuberante, incluyendo árboles, helechos y plantas con flores, todos dispuestos de manera intrincada en varios niveles conectados por senderos sinuosos y escaleras. Instalaciones artísticas, como estatuas y grandes pinturas abstractas, están colocadas cuidadosamente en todo el espacio, añadiendo capas de creatividad al entorno natural. Las personas exploran el área con calma, interactuando con el arte o simplemente disfrutando de la atmósfera serena. Un elemento central de agua atrae la atención, mientras que una mesa comunal ofrece un lugar para la reflexión tranquila o la conversación, integrando de manera armoniosa la naturaleza, el arte y la interacción humana. La combinación de elementos orgánicos y artísticos sugiere un lugar de intercambio cultural y contemplación.

When I worked as the interim General Manager for Interplanetary Shipyard, I collaborated with a team of engineers developing complex open-source protocols like IPFS and libp2p. To communicate effectively with them, I built custom ChatGPT models to help me interpret technical notes and ask clarifying questions. This allowed me to translate complex details into actionable steps during a deployment call with a client.

One memorable instance was during a call with a prospective client with an equally complex product. I quickly realized I was lost. Zero clue as to the technical problem being described or what to do about it. Cue: Panic.

Instead of melting down (or faking it), I used ChatGPT in real-time on the call to simplify technical notes, clarify constraints, and translate my understanding into actionable steps. Here's how:

Real-Time Interpretation Loop Triage

How I used ChatGPT to facilitate a technical deployment discussion with engineers and a client:

1. Input: Technical raw notes
   I provided ChatGPT with raw notes from the engineers, asking it to explain the project constraints in plain English.
   Output: A simplified summary that was easy for me to understand

2. Input: Clarifying questions

   I asked ChatGPT follow-up questions to clarify my understanding, such as, "Is this similar to other client requests?" or "Is this on our product roadmap?"
   Output: Clear, binary answers that helped me scope the problem

3. Input: Forward-looking questions
   Once I had a grasp of the issue, I asked ChatGPT, "Given these constraints, what are our options to move forward?"
   Output: A simple recommendation for next steps

4. Input: Simple recommendation
   I asked ChatGPT to help me translate the simple recommendation into technical language for our engineers.
   Output: I validated that the recommendation and translation made sense with the team, then posed the question in real time to the client on the call.

---

I have done some version of this exercise in many, many, jobs. Across many, many different types of people. It is helping me get better at being a business generalist, communicate my own ideas, and recognize areas where I need to bring in deeper, more specialized support. While it's not a perfect catch-all, tighter interpretation loops help me do something so critically important in any job: Keep forward momentum.

Whether you're a general manager trying to convey ideas to a technical team, a designer trying to share your creative vision with the CEO, or a job candidate trying to parse your career narrative for a hiring manager, figuring out to use AI to optimize your own interpretation loops is one of the biggest optimization opportunities for anyone in the working world today.

When applied correctly, tighter interpretation loops can reduce misunderstandings, streamline communication, and maintain momentum at work, even on projects that tend to get bogged down. It also makes it easier than ever to bring together collectives of people onto diverse teams, with many different backgrounds, communication styles, and even languages. This is one of the things I'm most excited to explore with AI-powered tools—how they can break down barriers and foster more effective collaboration across diverse, multi-lingual, and multi-disciplinary teams.