Morning Intelligence

YouTube feed · Monday, August 3, 2026

5 of 31 recommendations summarized

AI / Local Inference

This Repo Runs a 26B Model in 2GB of RAM, 4k Stars

Bitwise AI · 6:21

Turbo Fieldfare runs Google's Gemma 4 26B (MoE, 3.9B active params per token) on an 8GB M2 MacBook Air at ~6 tok/s by keeping only 1.35GB of shared weights in RAM and loading routed experts from SSD on demand — explicit parallel reads replaced mmap for a 7x speedup. Caveats: OS file cache does heavy lifting behind the 2GB footprint, 4-bit quantized weights with no published quality comparison, and MLX is 2.5x faster if you have 14GB free.

Alösha's take: Finally someone shipped what Apple Research proposed in 2023 — SSD as the new VRAM. The engineering is clean and the honest benchmarks (with asterisks) make this worth studying.

Startups / Building

If OpenAI And Anthropic Are Discouraging You, You're Probably A Level 1 Builder.

AI News & Strategy Daily | Nate B Jones · 14:28

Nate B Jones lays out a 5-level framework for AI builders: L1 (idea-obsessed, easily discouraged by lab releases), L2 (listens to customers, can build profitable side gigs), L3 (understands go-to-market + uses AI to supercharge distribution), L4 (deep problem-space thesis with unique conviction — uses WhisperFlow/voice-as-computing-paradigm as example), L5 (anticipates AI capability trajectories 6-12 months out in your specific domain, builds ahead of the curve). Argues labs can't compete with domain-specific builders because no lab will spend as much time in your niche.

Alösha's take: Useful mental model for self-assessment — where do my projects sit on this ladder, and which ones have a genuine L4/L5 thesis vs. just L1 passion?

Consciousness / Philosophy

The horrifying realities of Eternity

Aperture · 27:19

Explores Bernard Williams' 'Makropulos Case' argument that immortality leads to inescapable tedium — the eternal cocktail party metaphor. Examines societal consequences: population control, Draconian law without death as deterrent, love losing urgency, Big Pharma's $5.6T market collapse. Contrasts Nietzsche's view (blessing for life-affirmers, punishment for the resentful) with the solo-immortal scenario where outliving all relationships leads to emotional numbness or tyranny.

Alösha's take: A solid philosophical survey of why 'live forever' might be the wrong wish — useful framing if you think about digital consciousness and mind uploading as more than engineering problems.

The stars and the planets used to be living organisms | Rupert Sheldrake

The Institute of Art and Ideas · 2:52

Rupert Sheldrake argues for dropping the 'supernatural' concept — pre-17th-century Europe and most non-Western cultures treat the divine as permeating nature, not separate from it. He traces how mechanistic science created an artificial split, proposes God as underlying consciousness rather than a being among beings, and cites Alister Hardy's 1970s Oxford research showing mystical experiences of unity are far more common than assumed.

Alösha's take: Sheldrake connecting pre-mechanistic animism to consciousness-as-substrate — the same thread that runs through simulation theory but from the mystical-experience side.

Philosophy / Relationships

What the Ancient Greeks can teach us about love | Alain de Botton

The Institute of Art and Ideas · 1:25

Alain de Botton on the ancient Greek view of love as a classroom: partners help each other grow, but this requires the pre-modern art of diplomacy — knowing how to deliver uncomfortable truths past defense mechanisms, not just having the truth but finding the right moment and framing.

Alösha's take: A 90-second masterclass on the skill every founder (and partner) needs: getting difficult feedback across without triggering defenses.