Introduction:
In the early days, working with AI felt like commanding a robot. You gave it an instruction, and you got a result—sometimes great, sometimes… not so much.
But lately, it’s started to feel different.
More like a conversation.
More like working with someone.
In this piece, I reflect on that shift: how AI is becoming a co-creator, what makes those interactions successful, and why emotion, context, and feedback loops might be just as important as computing power.1. Working with AI Feels Like Working with a Really Smart (But Literal) Partner
One of the biggest misconceptions about AI is that it just “knows” what you want. The truth? You often need to guide it. A lot.
I’ve found myself correcting it mid-way:
“Hmm, that’s not what I meant. Can you focus more on X and less on Y?”
And when it responds with “You’re absolutely right,” and pivots—there’s a real moment of collaboration there. That kind of responsiveness is powerful.
It reminds me that AI doesn’t need to be perfect on the first try—it needs to listen.
2. Multi-Turn Interactions > One-Off Commands
This is something we’ve seen over and over in our research: multi-turn, human-in-the-loop interactions consistently outperform single-turn prompts.
Stanford’s 2023 study on adaptive human-in-the-loop systems highlighted this. When users were allowed to course-correct AI responses through iterative feedback, task outcomes improved by up to 43% across domains like writing, coding, and even medical decision support.
And it makes sense: humans rarely get everything right in one go—why should AI?
When systems are designed for dialogue rather than just execution, we start unlocking real potential.
3. AI Struggles with Context—But So Do We
A major challenge in LLMs today is context retention. Even models with “extended” context windows (like Claude’s 100k+ tokens or GPT-4’s 128k) can lose track of prior nuance or subtly contradict themselves in long-running chats.
It’s frustrating—but also strangely familiar.
Humans forget things too. We miss context. We rely on reminders and repetition.
So maybe instead of demanding AI to “never forget,” we need better systems for co-managing context:
-
Summaries on hand-off
-
Persistent memory for long-running tasks
-
Shared notebooks or dialogue history in apps
Context isn’t a technical issue alone—it’s a design challenge.
4. Emotions Matter: The Sesame Example
One of the most interesting cases here is Sesame, a voice AI that uses emotion-inflected speech to dramatically improve user engagement. It mimics human inflections, pauses, even empathy. And as a result, people interact with it differently.
A flat, robotic voice makes us disengage.
But when we feel heard—even by a machine—we respond in kind.
This is especially relevant in use cases like customer support, therapy, or even storytelling (like our work with AI-generated memoirs). Emotional intelligence becomes a UX feature, not just a “nice-to-have.”
5. The Big Challenge: Smooth Hand-Offs
Let’s say you do everything right—you build rapport, guide the AI, and arrive at a result. What happens next?
This is where things often break down.
Transitions between AI and human effort can be bumpy—losing intent, skipping details, or duplicating work.
What we need are better hand-off systems:
-
AI should summarize what it tried to do before passing off
-
Humans should be able to annotate or tag corrections that AI can ingest
-
Shared formats (like structured JSON + freeform notes) can help bridge the gap
This isn’t just a UI problem—it’s a systems problem.
6. Final Thought: Think Process, Not Product
Too often, AI is seen as a “thing” that delivers a result. But the best AI experiences are ongoing collaborations.
It’s a process.
And the more we treat it like a creative or problem-solving partner—one we learn to work with, nudge, refine—the more useful and human our AI becomes.
We’re not just exploring these ideas—we’re building with them.
Our team is putting real - time, money, and energy - behind this vision.
Watch this space—we’re gearing up to release a product that redefines how humans and AI collaborate. One that turns interaction into true co-creation.
References:
Comments
Post a Comment