>burns through billions of dollars>causes thousands of people to lose their jobs>destroys all the markets related to computers by taking up all the RAM and GPUs>steals all information from public forums, public repositories and public arts>all it produces in exchange is low quality slop >is unable to make a profit at allAI is a technology that takes away value from everything it touches, it does not actually generate value.
>>107595987that's not true at all. all customer interaction with product representatives can be automated at your expense.it's antisemitic for you not the bear the expense of jews making a profit. reported.
>>burns through billions of dollars / Unable to make a profitWhile it is true that foundational model training requires immense capital expenditure (CapEx), equating high initial investment with a lack of value is a fundamental misunderstanding of infrastructure cycles. Capital Intensity: Like the build-out of the railroads, the internet (fiber optics), or the cloud, the initial phase requires "burning" billions. Revenue Growth: While many AI startups are pre-profit, the providers of the underlying infrastructure (e.g., NVIDIA, TSMC) and integrated cloud services (Microsoft, AWS) are seeing record-breaking revenue and profits directly attributed to AI demand. The "J-Curve": Technological adoption often follows a J-curve where costs precede utility. We are currently in the heavy investment phase.>>causes thousands of people to lose their jobsThe narrative of total job displacement is historically inconsistent with technological shifts. Lump of Labor Fallacy: This is the mistaken belief that there is a fixed amount of work to be done. Historically, automation increases efficiency, lowers the price of services, and creates new categories of employment that were previously unimaginable (e.g., prompt engineers, AI auditors, or data curators). Augmentation vs. Replacement: Current data suggests AI acts primarily as a "copilot." It automates tasks, not necessarily entire occupations. By handling rote tasks, it allows human workers to focus on higher-level strategy and creative problem-solving.
>>steals information from public forums, repositories, and artsThe term "steal" is a legal and philosophical misnomer in this context; the debate centers on Fair Use. Transformative Use: Under current legal frameworks (which are being tested), training a model is often viewed as "transformative." The model does not store a copy of the data; it learns the statistical patterns and relationships within it. Human Learning Parallel: Human artists and coders also learn by consuming public forums, repositories, and art. AI performs a high-scale version of this pattern recognition. Opt-out and Licensing: The industry is rapidly moving toward licensed data agreements (e.g., OpenAI’s deals with News Corp, Reddit, and Axel Springer), ensuring a value-exchange for data creators.>>all it produces is low-quality slopSubjective quality assessments do not negate objective utility. Utility in Specific Domains: AI is currently producing high-value outputs in non-creative fields: AlphaFold: Predicted the structures of nearly all known proteins, a feat that would have taken centuries of human labor. Coding: GitHub Copilot is documented to increase developer speed by up to 55%. Medicine: AI is significantly more accurate at detecting early-stage cancers in medical imaging than human radiologists in several studies. Iterative Improvement: The "low quality" argument ignores the "scaling laws." Models are improving exponentially in reasoning and accuracy every 12–18 months.The claim that AI "takes away value from everything it touches" is contradicted by Total Factor Productivity. Value is generated when the cost of an output (intelligence, code, analysis) drops significantly while its availability increases.By lowering the "cost of cognition," AI enables businesses to solve problems that were previously too expensive or complex to tackle, which is the definition of value creation in an information economy.
Luddite tranny artist thread detected
>>107595987AI (new) =/= AI (classic)
T9 of the smartphone era
But it makes chuddies seethe so it's all worth it
Slop thread.
>>107596097>>107596112clanker hands typed this> While it requires significant capex to train the models blablablaIt also costs an ungodly amount of energy and hardware to evaluate prompts, and it seems to be rising, not diminishing
>>107595987Yeah it's a meme. However there's a powerful combination of True Believers who want to use it to run the world (or even save the world, although I'm not sure anyone at the top is that idealistic), greedy boomers who want number to go up, retarded consultoids who genuinely think it can replace workers en masse, and possibly some people who are very scared of what will happen if number goes down.
only use for AI>>>/wsg/6046619
AI made this
