scaling laws
30 articles about scaling laws in AI news
Scaling Laws Differ for Native Multimodal VLMs
A systematic study reveals distinct scaling laws for native multimodal pre-training, showing vision-language models require different compute-optimal size/token ratios than language-only models.
Roman Yampolskiy: 'AGI is a Question of Cost, Not Time' as Scaling Laws Hold
AI safety researcher Roman Yampolskiy argues that achieving AGI is now a matter of computational and financial resources, not theoretical possibility, citing the continued validity of scaling laws and early signs of recursive self-improvement.
UniMixer: A Unified Architecture for Scaling Laws in Recommendation Systems
A new arXiv paper introduces UniMixer, a unified scaling architecture for recommender systems. It bridges attention-based, TokenMixer-based, and factorization-machine-based methods into a single theoretical framework, aiming to improve parameter efficiency and scaling return on investment (ROI).
Dyna-2 World-Action Model Trained on 1M Hours Video
Dyna-2, trained on 1M+ hours of egocentric video, jointly predicts future video and actions. Claims new scaling laws but no benchmarks or technical details released.
Beyond the Benchmark: New Model Separates AI Hype from True Capability
A new 'structured capabilities model' addresses a critical flaw in AI evaluation: benchmarks often confuse model size with genuine skill. By combining scaling laws with latent factor analysis, it offers the first method to extract interpretable, generalizable capabilities from LLM test results.
Kimi Team's 'Attention Residuals' Replace Fixed Summation with Softmax Attention, Boosts GPQA-Diamond by +7.5%
Researchers propose Attention Residuals, a content-dependent alternative to standard residual connections in Transformers. The method improves scaling laws, matches a baseline trained with 1.25x more compute, and adds under 2% inference overhead.
Huawei's τ Scaling Law Redefines Transistor Race Without EUV
Huawei's τ Scaling Law at IEEE ISCAS replaces geometric transistor scaling with time-based optimization, targeting 1.4nm density by 2031 without EUV, challenging US export controls.
Lloyds Banking Group Details 'Atlas' ML Platform for Scaling AI in a
A technical blog post details how Lloyds Banking Group rebuilt its internal Machine Learning platform, Atlas, on a cloud-native architecture to overcome scaling limits and meet stringent regulatory requirements. This is a blueprint for operationalizing AI in high-stakes, governed industries.
Scaling Law Plateau Not Universal: More Tokens Boost Reasoning AI Performance
Empirical evidence indicates the 'second scaling law'—performance gains from increased computation—does not fully plateau for many reasoning tasks. Benchmark results may be artificially limited by token budgets, not model capability.
Agent Harness Scaling: EFC Predicts Success at R2 0.99 vs 0.42
New research introduces Effective Feedback Compute (EFC), which predicts agent success at R2 0.99 vs 0.42 for raw tokens. Reallocating compute by EFC lifts success 3x at the same budget.
Robotics' Scaling Breakthrough: How SONIC's 42M-Parameter Model Achieves Perfect Real-World Transfer
Researchers have demonstrated that robotics can scale like language models, with SONIC training a 42M-parameter model on 100M human motion frames. The system achieved 100% success transferring to real robots without fine-tuning, marking a paradigm shift in robotic learning.
Epoch AI: Parallelization limits could delay intelligence explosion
Epoch AI argues parallelization limits could delay an intelligence explosion by years, as scaling beyond 10^28 FLOP faces diminishing returns from hardware constraints.
Ahold Delhaize USA Scales Personalization Across Banners
Ahold Delhaize USA is scaling AI-driven personalization across banners like Stop & Shop and Giant Food, using data and ML to tailor shopping experiences. This matters for retail as it demonstrates a major grocer's commitment to AI for customer loyalty and revenue growth.
Anthropic Warns Upcoming LLMs Could Cause 'Serious Damage'
Anthropic has issued a stark warning that its upcoming large language models could cause 'serious damage.' The company states there is 'no end in sight' to capability scaling and proliferation risks.
US Data Center Power Demand Hits 15 GW, Grid Constraints Emerge
US data center power demand reached 15 gigawatts in 2023, up from 11 GW in 2022. This rapid growth highlights a widening bottleneck: compute infrastructure is scaling faster than power delivery systems can support.
daVinci-LLM 3B Model Matches 7B Performance, Fully Open-Sourced
The daVinci-LLM team has open-sourced a 3 billion parameter model trained on 8 trillion tokens. Its performance matches typical 7B models, challenging the scaling law focus on parameter count.
Gamma 31B Model Reportedly Outperforms Qwen 3.5 397B, Highlighting Efficiency Leap
A developer's social media post claims the Gamma 31B model outperforms the much larger Qwen 3.5 397B. If verified, this would represent a dramatic efficiency gain in large language model scaling.
QuatRoPE: New Positional Embedding Enables Linear-Scale 3D Spatial Reasoning in LLMs, Outperforming Quadratic Methods
Researchers propose QuatRoPE, a novel positional embedding method that encodes 3D object relations with linear input scaling. Paired with IGRE, it improves spatial reasoning in LLMs while preserving their original language capabilities.
The Two-Year AI Leap: How Model Efficiency Is Accelerating Beyond Moore's Law
A viral comparison reveals AI models achieving dramatically better results with identical parameter counts in just two years, suggesting efficiency improvements are outpacing hardware scaling. This development challenges assumptions about AI progress and has significant implications for deployment costs and capabilities.
Anthropic's RSP v3.0: From Hard Commitments to Adaptive Governance in AI Safety
Anthropic has released Responsible Scaling Policy 3.0, shifting from rigid safety commitments to a more flexible, adaptive framework. The update introduces risk reports, external review mechanisms, and unwinds previous requirements the company says were distorting safety efforts.
The Billion-Dollar Blind Spot: Why AI's Evaluation Crisis Threatens Progress
AI researcher Ethan Mollick highlights a critical imbalance: while billions fund model training, only thousands support independent benchmarking. This evaluation gap risks creating powerful but poorly understood AI systems with potentially dangerous flaws.
Fchollet: Future AI Will Be 'Incredibly Cheap' to Train
Chollet claims frontier AI training will become incredibly cheap, challenging the assumption of permanent high costs. The post offers no evidence, making it a speculative provocation.
AI data centers could add 1.4°C to global warming by 2060, paper finds
AI data centers could add 1.4°C to global warming by 2060, per a new arXiv preprint, assuming 30% annual compute growth. The paper highlights the need for policy intervention.
OpenAI Launches GPT-5.6 Sol Under US Government Restrictions
OpenAI's GPT-5.6 Sol beats Claude Mythos 5 in agentic coding (88.8% vs 88%) but US government restricts access to select partners, a policy OpenAI calls unsustainable.
Jim Keller: Tenstorrent IPO Looms as BlackHole Chip Scales
Jim Keller confirmed Tenstorrent's IPO plans as BlackHole chip scales for AI inference, competing with Nvidia. No revenue disclosed.
Colossus 2: xAI's Memphis Cluster Hits 300,000 GPUs
xAI's Colossus 2 hits 300,000 GPUs, targeting 1M by year-end. Training Grok-3, the $6B cluster challenges OpenAI and Google.
Why Traditional Retail Metrics Break Down in Agentic Commerce
Valtech's 2026 research shows 96% of retailers face integration barriers, 48% are stuck in AI pilot purgatory, and nearly 75% can't link AI spend to metrics, as agentic commerce fragments customer journeys beyond traditional measurement frameworks.
11-Agent Company Earned $0: CLAUDE.md Mistakes Cost Revenue
11-agent company experiment earned $0 after 896 tasks. Operator open-sourced CLAUDE.md template with 72 lessons on coordination failures and legal constraints.
NHN Deploys 7,656-GPU AI Cluster in Seoul
NHN launched a 7,656-GPU cluster in Seoul, South Korea, for domestic enterprise AI workloads. The cluster targets inference and training, competing with Naver and Kakao.
Billionaire Sues Tiny Michigan Township to Force OpenAI Data Center Through
Billionaire Steven Roth's Related Digital sued Saline Township, Michigan, after it rejected a 21M sq ft OpenAI data center, forcing approval via 'exclusionary zoning' claim.