Pao Carreon

AI • SaaS • Startups

Sharing useful tools, growth strategies and tech insights.

Learning in public.

[Pao Carreon @1ncursio]

Hi, I'm Pao.

Interested in AI, SaaS, digital marketing and startups.

Using this account to document what I'm learning and share useful tools and insights.

Always open to connecting with builders and marketers.

[Pao Carreon @1ncursio]

Using the latest AI model feels a bit like chasing the meta in competitive games.

It can give you an edge.

But fundamentals still win.

Knowing how to ask good questions, verify answers and apply the information matters more than simply using the newest model.

[Pao Carreon @1ncursio]

The AI race isn't slowing down.

But it feels like the focus is changing.

Less:
"Which model has the highest benchmark score?"
More:
"Which AI system actually helps people get work done?"

Real-world usefulness is becoming the benchmark that matters most.

[Pao Carreon @1ncursio]

@aditabrm @micro1 @micro1_ai @reductoai That's why independently audited benchmarks matter. They provide a more realistic picture of how models and document processing systems perform on complex, long-form documents, where tradeoffs in precision, recall and completion become much more apparent.

[Pao Carreon @1ncursio]

@sid_mnk @Pulse__AI This reinforces the idea that AI performance isn't just about choosing the most capable model.

How information is extracted, structured and presented to the model can have a huge impact on the final result.

Better context often beats a bigger model.

[Pao Carreon @1ncursio]

@robert_lauko @KurateOrg @MatthiasHafner1 @NicolasOderbolz If AI can reliably help prioritize high-impact research, it could make it much easier for researchers to navigate the growing volume of scientific papers.

The key will be ensuring these rankings complement, rather than replace, expert review.

[Pao Carreon @1ncursio]

@danwilliamsphil It suggests the impact of LLMs depend less on their tendency toward sycophancy and more on how people use them. If they encourage users to consider alternative perspectives, they could reduce polarization rather than reinforce it.

Replication on different topics is interesting.

[Pao Carreon @1ncursio]

One thing AI has changed for me:

I ask more questions.

Instead of stopping when I don't understand something, I can dig deeper, ask follow-up questions, compare different explanations, and keep learning.

For me, AI isn't replacing curiosity—it's making it easier to satisfy it.

[Pao Carreon @1ncursio]

One underrated AI skill isn’t prompting.

It’s knowing when to question the answer.
The faster AI gets, the more valuable critical thinking becomes.

Using AI well isn’t just about getting responses, it’s about knowing what to trust and what to verify.

[Pao Carreon @1ncursio]

@perrymetzger That's the part that stands out to me too.

A machine-checkable proof isn't just about reaching the right conclusion, it's about making every step independently verifiable.

That kind of rigor could have applications beyond math, from software verification to scientific research.

[Pao Carreon @1ncursio]

@XFreeze Privacy is quickly becoming a competitive feature for AI platforms, not just a compliance checkbox.

Features like Zero Data Retention can make it easier for organizations to adopt AI for sensitive workflows while maintaining stronger data governance.

[Pao Carreon @1ncursio]

@SVVVAYED I like this analogy.

For me, AI isn't valuable because it's always right, it's valuable because it helps expose gaps in my own thinking.

Sometimes the best outcome is realizing why the AI's answer doesn't work.

[Pao Carreon @1ncursio]

AI is becoming less about replacing work and more about removing friction.

Whether it's brainstorming ideas, summarizing research or organizing information, even saving 15–20 minutes a day adds up over time.

Small improvements compound.

[Pao Carreon @1ncursio]

@OpenAI Interesting to see AI models becoming more specialized instead of chasing a one-size-fits-all approach.

Choosing between capability, efficiency and cost depending on the use case feels like the direction the ecosystem is heading.

[Pao Carreon @1ncursio]

@FoundationOAI As AI lowers the barriers to scientific discovery, investing in prevention and preparedness feels just as important as advancing the technology itself.

The goal should be to maximize AI's benefits while building safeguards that help society stay resilient.

[Pao Carreon @1ncursio]

@AnthropicAI Interesting to see AI models being treated as critical infrastructure capabilities.

As these systems become more specialized, access decisions will likely shape not just innovation, but also how governments and organizations approach cybersecurity.

[Pao Carreon @1ncursio]

@oziadias @Nature One of the most exciting possibilities is using AI to uncover patterns that humans might overlook.

If AI consistently identify signals linked to disease risk, it could help generate research questions, not just diagnoses.

That feels like where the biggest discoveries are coming

[Pao Carreon @1ncursio]

@calcsam This feels like one of those features users won't notice until something goes wrong.

Real-time persistence during streaming makes AI agents much more resilient to refreshes, disconnects and unstable networks. Small UX improvement, big impact on reliability.

[Pao Carreon @1ncursio]

I don't use AI to make decisions for me.

I use it as a second opinion.

Sometimes it challenges my assumptions.
Sometimes it confirms my thinking.
Either way, it helps me think more critically instead of less.

[Pao Carreon @1ncursio]

Using AI effectively is becoming a skill of its own.

The difference isn't which model you use, it's knowing how to ask better questions, verify the answers, and turn them into something useful.

The tools will keep evolving. The skill of using them well will matter even more.

[Pao Carreon @1ncursio]

AI feels like it's entering a new phase.

A year ago, most conversations were about generating text.

Now the focus is on AI agents that can actually complete tasks and fit into real workflows.

The next few years should be interesting to watch.