Paola Santiago
psproductperson.bsky.social
Paola Santiago
@psproductperson.bsky.social
Director of Product Management in Startups turned consultant and educator to make understanding tech more accessible for everyone.
Substack: https://psproductpersonpapers.substack.com/
Youtube: https://www.youtube.com/@psproductperson
Let me re-introduce myself. My name is Paola and I've worked to build apps and SaaS products for a long time. And now, I can build it for you. 

A SaaS product or customised software used to need 6 months. And now you can build the beta in 1 month. Check out my website: https://psproductperson.com/
August 27, 2025 at 3:04 PM
1 in 4 Americans would switch doctors if they don’t use AI. Wild, right?

But here’s the thing:
AI also hallucinates...and it learns from biased data.

In my new video:
How LLMs could harm real patients
Racial bias in pulse oximeters
What public trust in AI is really built on

#AI #AItrust
April 17, 2025 at 2:03 PM
Did you know AI learns how to sound human from Reddit posts? 

Here’s how ChatGPT picked up empathy, support, and emotional language from upvoted Reddit threads and why that matters.

This isn’t just tech, it’s how we’re shaping AI behavior without even knowing it.

#AIExplained #ChatGPT #AI
April 16, 2025 at 2:12 PM
We’ve been forming emotional bonds with machines since the 60s. ELIZA didn’t understand language...yet people still trusted it.

Today? We’re doing the same with ChatGPT.
But AI doesn’t understand you. It just predicts.

So why does it feel so real?

#AIexplained #AI #AItherapy #TechAndSociety
April 15, 2025 at 2:02 PM
Would you rather talk to AI than a therapist?

1 in 4 Americans say yes.

AI isn’t just being used as a tool, it’s being seen as a listener. And in some cases, a more empathetic one than humans.

I explore why so many people are pouring their hearts out to language models like ChatGPT.

#AI
April 14, 2025 at 2:03 PM
Why does ChatGPT give you a different answer than me... even with the same question? 
 
Let’s break down AI memory, context windows, and how language models like GPT-4 make you feel heard. Is it real or just next-word prediction? 

#chatgpt #aiexplained #gpt4 #llm #artificialintelligence
April 10, 2025 at 2:04 PM
Ever feel like ChatGPT knows you?

This video breaks down how memory, context windows, and prediction create the illusion of connection. If you're a creator, researcher, or builder trying to understand how people fall for AI, this is for you.

#learnai #creatorinsight #chatgptexplained #ai
April 9, 2025 at 2:03 PM
Why do people feel emotionally connected to ChatGPT?

This video explains how AI seems to understand you... even though it doesn’t.

Perfect for anyone curious about AI relationships, emotional AI, and the psychology behind human-AI interaction.
#chatgpt #airelationship #emotionalai #ai
April 8, 2025 at 2:03 PM
AI doesn’t just pass the Turing Test
It holds conversations, remembers what you like, and mirrors your emotions. In just a few years, we’ve gone from clunky Facebook bots to AI that feels almost human.

How do we feel about this development? 
#AI #ArtificialIntelligence #TuringTest
April 7, 2025 at 2:06 PM
With my ongoing obsession to understand social behaviour and technology.

I wanted to explore the AI mechanics: memory through context window. This alone makes it feel like ChatGPT remembers our preferences and who we are.

What do you think about that?
April 2, 2025 at 10:02 PM
AI has passed the Turing Test. It can hold conversations, remembers what you like, and mirrors your emotions.

So what are ways that we're building to safeguard ourselves from the future of AI?

#AI#ArtificialIntelligence#TuringTest#ChatGPT#EmotionalAI
March 30, 2025 at 3:03 AM
Gen Z isn’t Googling anymore.
They’re using ChatGPT, TikTok, and Instagram to search.
As a millennial, this blew my mind.
So I asked: why?

#AI #GenZ #SearchBehavior #TechTrends #UX
March 29, 2025 at 7:02 AM
Half of online traffic isn’t human. A third of it? Bad bots. And AI is making them smarter.

Bots don’t just spam anymore. They mimic real conversations, generate deepfake profiles, and bypass CAPTCHAs. If you’re relying on engagement metrics, how much of it is even real?
March 18, 2025 at 5:04 PM
AI isn’t always right, it can make things up.

In 2023, a lawyer used ChatGPT to draft a legal brief. The AI fabricated cases, quotes, and citations, leading to court sanctions. He never fact-checked.

Don’t let AI hallucinations fool you. Get the guide: https://ps-product-person.kit.com/2d98046718
March 11, 2025 at 7:04 PM
AI isn’t neutral, it reflects its training data.

In 2018, Amazon scrapped an AI recruiting tool that penalized resumes from women because it learned bias from past hiring data.

How does AI bias happens & how PMs can prevent it? Get the PMs guide here: https://ps-product-person.kit.com/2d98046718
March 10, 2025 at 7:03 PM