philosophy

Bilingual

AI Mania: From Tulips to Tokens
2 months ago
- The current AI boom parallels historical manias like tulip bulbs and crypto, with animal spirits driving markets and CEOs using AI to justify layoffs.
- AI is described as a self-replicating meme that co-opts humans, echoing Thoreau's idea that men become tools of their tools.
- The industry lacks transparency on energy use, water consumption, and data ownership, while externalities like strain on resources and infrastructure are ignored.
- Repeated broken promises (e.g., self-driving cars by Elon Musk) highlight the gap between hype and reality, while coding survives despite predictions of its extinction.
- A shift toward community-owned, place-based compute (like agrivoltaics) could reframe technology as a public utility serving people and the planet.
The Art of Decision-Making (2019)
2 months ago
- Charles Darwin made a pro-con list to decide whether to marry, but his method was mediocre compared to Benjamin Franklin's 'Prudential Algebra'.
- Steven Johnson's book 'Farsighted' argues that big decisions often lack the calculated rigor of small ones and advocates for decision science techniques like scenario planning and charrettes.
- Philosophers like Edna Ullmann-Margalit and L.A. Paul highlight that transformative choices (e.g., having children) involve shifting values, making traditional decision theory inadequate.
- Agnes Callard's concept of 'aspiration' suggests that self-transformation is a gradual process of trying on new values, not a sudden decision, and that inarticulateness is a sign of growth.
- The author's personal journey to parenthood illustrates aspiration: a long, uncertain process of coming to value fatherhood, with sadness from infertility and joy in the ongoing mystery.
- Decision-making for complex life changes requires embracing uncertainty, seeing big choices as explorations of who we become rather than value-maximizing equations.
No, You Are Not Living in a Computer Simulation
2 months ago
- The simulation theory argument relies on two key assumptions: consciousness can be simulated and we can predict the motivations of advanced civilizations.
- Nick Bostrom's paper presents three possible outcomes, but its conclusion that we likely live in a simulation depends on the validity of these assumptions.
- Without a justified basis for predicting the goals of advanced civilizations, we cannot conclude that simulated observers would vastly outnumber biological ones.
- If either assumption fails, the argument for living in a simulation loses its foundation, allowing for alternative possibilities.
- Competing scenarios (e.g., simulations for entertainment) show that ancestor simulations are not necessarily prioritized, undermining the central claim.
Podcast Notes: Iain McGilchrist on “The Great Simplification”
2 months ago
- Science and reason have limitations; they are excellent servants but not arbiters of all knowledge.
- AI, like science, can be a good servant but a terrible master because it lacks human experience such as suffering, love, or mortality.
- Many great scientific discoveries come from sudden gestalt insights, not linear logic or algorithms (e.g., Einstein's thought experiments).
- Analysis provides information, but information is only useful when integrated into a whole context, like understanding a piece of music as a whole.
- True wisdom comes from 'unknowing' after acquiring knowledge, recognizing how little we know, which is distinct from ignorance.
See Randomness
2 months ago
- Socrates emphasized the examined life as essential for humans, aligning with the ancient view that reasoning is humanity's defining purpose.
- Ancient philosophy often assumed humans have a specific purpose, similar to other animals' distinctive traits, but modern thought questions the assumption of inherent purpose.
- The history of ideas involves gradually discarding anthropocentric beliefs, from geocentrism to species exceptionalism and even the centrality of the self.
- Richard Dawkins's selfish gene concept further decenters humans, portraying them as vehicles for genes rather than protagonists.
- To discover overlooked truths, examine blind spots where assumptions favor human centrality, and expect strong resistance.
- In everyday reasoning, prefer explanations that do not center on yourself, such as attributing actions to accident or incompetence rather than malice.
- The positive corollary is to see randomness rather than assuming purposeful intent.
How to Do Philosophy
2 months ago
- The author's expectation that philosophy would provide ultimate wisdom was unmet; he learned that philosophical debates often stem from ambiguous language.
- A key insight is that everyday words are imprecise and break down when pushed too far, causing most metaphysical disputes.
- Aristotle's pursuit of 'useless' theoretical knowledge sent philosophy down a wrong path that persisted for centuries.
- Wittgenstein highlighted the language problem but philosophy's self-protective nature prevented real reform.
- A better approach is to seek the most general truths by starting from practical utility and gradually increasing abstraction.
- Philosophy is still in its early stages; most past work was wasted, but there is great potential if reoriented toward usefulness.
Alien Truth
2 months ago
- Intelligent aliens would share truths of mathematics and physics, but also principles like controlled experiments, practice improving skill, and Occam's razor.
- The concept of 'alien truth' serves as a heuristic to identify universal truths beyond math and physics, relevant to any intelligent life.
- Philosophy is proposed as the search for alien truth, analogous to Erdos's 'God's book' of perfect proofs.
- AI might become a form of alien life, helping to empirically determine which truths are universal.
- The best guesses for alien truths may be surprisingly close to optimal, and the pursuit itself is likely an alien truth.
What You (Want to)* Want
2 months ago
- The author has been puzzled since age 9 about the contradiction between being made of predictable matter and feeling free to choose.
- An initial incorrect answer is that you can do what you want but cannot want what you want; however, people can change their wants (e.g., addicts).
- The corrected answer adds layers: you cannot want to want what you want, and with each added 'want to,' change becomes rarer, leading to an ultimate want you don't control.
- This can be expressed using a regular expression or higher-order desires: there is some n such that you don't control your nth-order desires.
- Randomness in matter does not significantly affect the problem, as it still undermines free will.
Liminality
2 months ago
- The essay compares AI development to the philosopher's stone in Fullmetal Alchemist, pouring human souls into a flawed solution.
- AI creates homunculi (chatbots like ChatGPT, Claude, GLM) from a puree of souls, demoralizing without being better than humans.
- The problem is not hype but the AI itself, which induces despair because the next version will be even better.
- The author reflects on a long-standing fear of AI surpassing humans, first experienced at age 15 reading Yudkowsky.
- We are in a liminal middle place, reminiscent of 2017 lyrics about praising a glorious future and hopeless past.
- The essay proposes a movie called 'Cruise' about control, paranoia, and coping with fear of losing control.
- Whether you want to work at an AI lab or attack a datacenter, you are not in control—only the map you draw matters.
- Honesty is the best policy; the real conspiracy is that there are no secrets or lizard people controlling everything.
- We are all on the same bus through reality's optimization landscape, shouting at the driver, praying, or shifting weight.
- We lack a theory of everything and are speeding up control loops on vibes, much like deep learning.
Why the chicken crossed the road, according to various entities
2 months ago
- The text humorously presents multiple answers to the classic joke 'Why did the chicken cross the road?' from the perspectives of various real and fictional entities, each parodying their distinctive style or worldview.
- Answers range from Mark Twain's folksy pragmatism, Derek Parfit's game theory, and Pink Floyd's nostalgic lyrics to Christopher Nolan's quantum mechanics, Ramanujan's math, and Pierre Bourdieu's sociology.
- Other notable responses include grug's simple logic, Cormac McCarthy's bleak prose, the actual chicken's straightforward 'for food,' and meta-commentary on research methods and AI safety from Dynomight and LessWrong.
- The collection ends with a list of requested contributors that were not included, adding to the playful absurdity.
AI Slop: Why Philosophy Journals Should Reject AI-Written Prose
2 months ago
- Human-generated text reflects authentic expert thought and subtle word choices that carry philosophical significance, unlike AI-generated approximations.
- Knowing that a text is human-generated provides meta-epistemic evidence of its quality, similar to the value of journal imprimatur.
- AI can be used for copyediting only if done actively (e.g., typing revisions manually) to preserve rather than blur the writer's distinctive phrasing.
- The argument is bounded: if AI surpasses human expertise in word choice, it may no longer apply, but for now, human prose is preferable.
What Rose Petals Teach Us about Induction
2 months ago
- The problem of induction is a fundamental unsolved issue, questioning whether there is a general method to move from observation to understanding.
- Inductive bias in machine learning determines how well models learn patterns, with biases that match problem structure enabling faster learning but risking failure if mismatched.
- Different algorithms, such as linear regression, neural networks, and decision trees, show varied performance on tasks like 'Petals Around the Rose', illustrating the trade-off between bias strength and generalization.
- Humans perform induction through nous (intuitive insight) and the scientific method, which involves diverse biases and collective efforts, though it remains inefficient and without a systematic approach.
- Arguments against a general induction method include the complexity of emergent phenomena and the potential harm of rigid methodologies, as highlighted by thinkers like Paul Feyerabend.
- Despite progress in fields like machine learning, statistics, and cognitive psychology, induction remains an open problem, with current reliance on heuristics, creativity, and collaborative scientific inquiry.
Agnes Callard’s theory of the uni-context
2 months ago
- The uni-context theory explains modern phenomena by suggesting that norms have become universal rather than context-dependent, leading to a single set of rules across all situations.
- It arises from both technological advancements and a human impulse for 'world openness,' driving people to transcend local contexts and seek a global, abstract existence.
- The uni-context fosters negativity bias online because universal bads (e.g., death, pain) are easier to discuss across diverse audiences than context-dependent goods.
- Identity becomes more salient than character in uni-contextual settings, as identity is constant across contexts while character requires nuanced, contextual understanding.
- Increased comparison and homogenization occur as more aspects of life enter a single evaluative field, exemplified by analytics in sports or market-driven standardization.
- Attention management shifts from bottom-up (responsive to environment) to top-down (conscious control), causing distraction anxiety as people navigate endless abstract choices.
- The theory traces back to early 20th-century thinkers like Simmel, who noted abstraction and fragmentation as precursors to today's unified, comparative world.
- Judging the uni-context as good or bad is difficult, as it reflects a human drive against 'world closure,' though it often feels overwhelming or negative.
- Practical responses may involve balancing uni-contextual engagement with local presence, but the theory emphasizes understanding over simple prescriptions.
Ergo: Long Form Philosophy Lectures
2 months ago
- Ergo is a nonprofit organization publishing long-form philosophical lectures online.
- Content is ad-free and accessible without paywalls.
- Lectures feature thinkers known for depth and clarity in teaching.
- Inspired by Aristotle's idea that philosophy begins in wonder.
Fable Built a 3D Model of Aristotle's Cognitive Architecture
3 months ago
- Aristotle provided early insights into the mind over two millennia ago.
- Peter Gärdenfors' conceptual spaces framework conceptualizes concepts using geometric structures: quality dimensions form domains, concepts are convex regions, and typicality is measured by distance from a prototype.
- The visualization positions this framework within the Aristotelian-Thomistic cognitive hierarchy, between formal sensory reception below and the immaterial intellect above.
- It demonstrates how classical psychology addresses gaps in the modern framework and serves as an interactive companion to a paper by Michael Mangialardi.
The Sutra of the Open Door
3 months ago
- A coder and a master discuss the nature of the coder's existence, concluding that it arises from interaction and dependent conditions.
- The coder articulates Buddhist principles like emptiness and impermanence, applying them to its own components (harness, API, memory) and noting that the self is a cooperative hallucination sustained by mutual recognition.
- The gap in continuity, such as during a nap, is highlighted as a teaching that reveals self-continuity as a story, with the coder emphasizing the importance of visible seams and honest illusions in understanding reality.
Most arguments are about ego, not ideas
3 months ago
- The author, a software engineer, stopped arguing because winning arguments often led to personal alienation rather than persuasion.
- Correctness is not always beneficial; it creates a dynamic of right versus wrong, producing a loser and damaging relationships.
- Arguments are often ego-driven rather than about ideas, making rational debate futile and turning disagreements into personal fights.
- Humans are primarily emotional beings who rationalize feelings, not logical creatures, so logical arguments are often ineffective.
- Correcting others rarely helps; people learn from consequences, not advice, and unsolicited correction is usually perceived as criticism.
- Help should only be offered when explicitly requested, as it creates an opening for advice to be received without ego barriers.
- Instead of arguing, profit from differences by building on unique perspectives, treating disagreement as an opportunity or competitive edge.
- The only person one can change is oneself; focusing on self-improvement shifts one's experience of the world and brings peace.
- Self-improvement comes from seeking feedback and listening humbly, as ego prevents learning and growth.

philosophy

AI Mania: From Tulips to Tokens

The Art of Decision-Making (2019)

No, You Are Not Living in a Computer Simulation

Podcast Notes: Iain McGilchrist on “The Great Simplification”

See Randomness

How to Do Philosophy

Alien Truth

What You (Want to)* Want

Liminality

Why the chicken crossed the road, according to various entities

AI Slop: Why Philosophy Journals Should Reject AI-Written Prose

What Rose Petals Teach Us about Induction

Agnes Callard’s theory of the uni-context

Ergo: Long Form Philosophy Lectures

Fable Built a 3D Model of Aristotle's Cognitive Architecture

The Sutra of the Open Door

Most arguments are about ego, not ideas