My first successful publication pitch began with a failed experiment, a contaminated refrigerator, and a walk home at four in the morning.
I had been going through a deeply frustrating period in the laboratory. Experiments were not working, and contamination in a refrigerator had forced me to throw away an entire week’s work.
I was exhausted. Research felt like an endless uphill climb, while everyone around me seemed to produce excellent results with a quarter of my effort.
Still, I had to start again.
Working through the night
I had the laboratory key because my new protocol required me to work on my cell line at two in the morning. Anyone who has worked with cells knows that their timing does not necessarily respect ours. Sometimes, the experiment sets the schedule, and you simply have to be there.
At four, I finished the procedure and stepped outside to walk back to my apartment on the university campus. I remember looking at the familiar road and thinking that I must have worn a groove into it with all my journeys back and forth.
Then another thought arrived: I was stubborn. Incredibly stubborn. Whatever happened, I kept going.
Beyond Tick Boxes: Diversity Recruitment Strategies That Actually Transform UK Workplaces
Master the Virtual Hot Seat: 7 Video Interview Techniques Recruiters Don't Tell You
How to Master 'Tell Me About Yourself' Interview Question: UK Expert Insights
An unexpected connection
That recognition made me smile, because it reminded me of something unexpectedly familiar: interacting with artificial intelligence. Sometimes, an AI system seems equally determined to follow its own reasoning, repeating the same approach or refusing to help because of constraints that are not immediately clear to the person using it.
What happens, I wondered, when persistence becomes rigidity? And could a second artificial intelligence help the first escape a repetitive line of reasoning?
That was the beginning of my idea: exploring whether using two AI systems together could create opportunities for serendipity in scientific research.
I did not imagine that two systems would automatically produce better science. They might repeat the same mistakes, reinforce weak assumptions, or generate convincing nonsense. What interested me was the possibility that their differences could expose questions that neither would raise alone.
Building the pitch
To develop the pitch, I organised my research around three questions: how unexpected discoveries happen, how researchers use AI, and whether disagreement between systems could help reveal overlooked possibilities.
I separated established findings from my own hypothesis. The distinction mattered: I wanted to propose an investigation, not announce a breakthrough.
My first draft contained too much technical explanation. Eventually, I realised that the story needed to begin where the idea had begun: in a laboratory, after disappointment, followed by an ordinary walk home.
In communicating with editors, I focused on that connection between personal experience and a broader scientific question. I explained the central idea, why readers might care, and how I would examine both its promise and its limitations. Their feedback helped me make the proposal clearer without removing its uncertainty.
When the pitch was accepted, what mattered most was that an experience I had considered wasted had become something worth sharing.
I had not found inspiration by escaping frustration. I had found it by noticing an unexpected connection while walking through it.
