C:\DINOKI\BLOG\INTRODUCING-OSAURUS-PERSONAL-AI-REIMAGINED-FOR-APPLE-SILICON.TXT

[Product Announcements]August 25, 2025•3 min read

> Introducing Osaurus: Personal AI, Reimagined for Apple Silicon

Today we're launching Osaurus — a new open-source AI server built natively for Apple Silicon. At just 7 MB, Osaurus makes it effortless to run modern language models on your Mac, with speed, simplicity, and privacy at the core.

D

Dinoki Team

READ_MORE >

C:\DINOKI\BLOG\WHY-BUILD-LOCAL-FIRST-APPS.TXT

[Product Philosophy]August 13, 2025•4 min read

> Why Build Local-First Apps

We spend hours each day on our computers—especially those of us working in digital fields. In many ways, we're citizens of the connected world. AI is reshaping our tools and workflows at an incredible pace.

D

Dinoki Team

READ_MORE >

C:\DINOKI\BLOG\DESIGNING-FOR-ARTIFICIAL-EMPATHY.TXT

[AI Ethics & Design]July 12, 2025•8 min read

> Designing for Artificial Empathy

In Snow White, the Queen turns to her Magic Mirror and asks, "Who's the fairest of them all?" She's not seeking information. She's seeking affirmation. And yet the Mirror answers coldly — as if facts could satisfy longing. We're now building mirrors of our own.

D

Dinoki Team

READ_MORE >

C:\DINOKI\BLOG\PEOPLE-SPIRITS-EXPLORING-EMOTIONAL-AI-THROUGH-AUTONOMOUS-COMPANIONS.TXT

[AI Research]June 23, 2025•6 min read

> People Spirits: Exploring Emotional AI Through Autonomous Companions

As Andrej Karpathy described them, LLMs are "people spirits" — stochastic simulations of human behavior embodied in autoregressive transformers. This isn't mere anthropomorphism; these models exhibit what can only be called an emergent psychology, shaped by the vast corpus of human expression they've absorbed.

D

Dinoki Team

READ_MORE >

C:\DINOKI\BLOG\UNLOCKING-COMMUNITY-TRUST-PRIVACY-FIRST.TXT

[Privacy & Security]June 14, 2025•8 min read

> Unlocking Community Trust: How Dinoki Puts Privacy First

When we set out to create Dinoki—a pixel-art AI companion that lives on your desktop—we faced a fundamental question: How do we build an AI sidekick that users can truly trust?

D

Dinoki Team

READ_MORE >