# From the Pursuit of Universal AGI Architecture to Systematic Approach to Heterogenous AGI: Addressing Alignment, Energy, & AGI Grand Challenges

arXiv: https://arxiv.org/abs/2310.15274 · HTML: https://mlsanity.com/paper/2310.15274/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 38 of its 221 references that are in the map.
- First posted on arXiv 23 Oct 2023.

## Builds on (most cited first, 30 of 38)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://mlsanity.com/paper/1502.03167/) (2015)
- [On the Opportunities and Risks of Foundation Models](https://mlsanity.com/paper/2108.07258/) (2021)
- [Finetuned Language Models Are Zero-Shot Learners](https://mlsanity.com/paper/2109.01652/) (2021)
- [Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification](https://mlsanity.com/paper/1502.01852/) (2015)
- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [Training Compute-Optimal Large Language Models](https://mlsanity.com/paper/2203.15556/) (2022)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [Emergent Abilities of Large Language Models](https://mlsanity.com/paper/2206.07682/) (2022)
- [Language Models (Mostly) Know What They Know](https://mlsanity.com/paper/2207.05221/) (2022)
- [Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour](https://mlsanity.com/paper/1706.02677/) (2017)
- [Energy and Policy Considerations for Deep Learning in NLP](https://mlsanity.com/paper/1906.02243/) (2019)
- [A General Language Assistant as a Laboratory for Alignment](https://mlsanity.com/paper/2112.00861/) (2021)
- [Show Your Work: Scratchpads for Intermediate Computation with Language Models](https://mlsanity.com/paper/2112.00114/) (2021)
- [BBQ: A Hand-Built Bias Benchmark for Question Answering](https://mlsanity.com/paper/2110.08193/) (2021)
- [ERNIE: Enhanced Language Representation with Informative Entities](https://mlsanity.com/paper/1905.07129/) (2019)
- [Are Emergent Abilities of Large Language Models a Mirage?](https://mlsanity.com/paper/2304.15004/) (2023)
- [A Comprehensive Survey of Neural Architecture Search: Challenges and Solutions](https://mlsanity.com/paper/2006.02903/) (2020)
- [Sustainable AI: Environmental Implications, Challenges and Opportunities](https://mlsanity.com/paper/2111.00364/) (2021)
- [PowerSGD: Practical Low-Rank Gradient Compression for Distributed Optimization](https://mlsanity.com/paper/1905.13727/) (2019)
- [TPU v4: An Optically Reconfigurable Supercomputer for Machine Learning with Hardware Support for Embeddings](https://mlsanity.com/paper/2304.01433/) (2023)
- [Bayesian Optimization is Superior to Random Search for Machine Learning Hyperparameter Tuning: Analysis of the Black-Box Optimization Challenge 2020](https://mlsanity.com/paper/2104.10201/) (2021)
- [Principles alone cannot guarantee ethical AI](https://mlsanity.com/paper/1906.06668/) (2019)
- [A Study and Comparison of Human and Deep Learning Recognition Performance Under Visual Distortions](https://mlsanity.com/paper/1705.02498/) (2017)
- [A Survey of Neuromorphic Computing and Neural Networks in Hardware](https://mlsanity.com/paper/1705.06963/) (2017)
- [TT-Rec: Tensor Train Compression for Deep Learning Recommendation Models](https://mlsanity.com/paper/2101.11714/) (2021)
- [Brain-inspired computing: We need a master plan](https://mlsanity.com/paper/2104.14517/) (2021)
- [The BrainScaleS-2 accelerated neuromorphic system with hybrid plasticity](https://mlsanity.com/paper/2201.11063/) (2022)

## Built on by (most cited first, 1 of 1)

- [Alignment and Safety in Large Language Models: Safety Mechanisms, Training Paradigms, and Emerging Challenges](https://mlsanity.com/paper/2507.19672/) (2025)

## Abstract (arXiv)

Artificial intelligence (AI) faces a trifecta of grand challenges: the Energy Wall, the Alignment Problem and the Leap from Narrow AI to AGI. We present SAGI, a Systematic Approach to AGI that utilizes system design principles to overcome the energy wall and alignment challenges. This paper asserts that AGI can be realized through multiplicity of design specific pathways and customized through system design rather than a singular overarching architecture. AGI systems may exhibit diver architectural configurations and capabilities, contingent upon their intended use cases. Alignment, a challenge broadly recognized as AIs most formidable, is the one that depends most critically on system design and serves as its primary driving force as a foundational criterion for AGI. Capturing the complexities of human morality for alignment requires architectural support to represent the intricacies of moral decision-making and the pervasive ethical processing at every level, with performance reliability exceeding that of human moral judgment. Hence, requiring a more robust architecture towards safety and alignment goals, without replicating or resembling the human brain. We argue that system design (such as feedback loops, energy and performance optimization) on learning substrates (capable of learning its system architecture) is more fundamental to achieving AGI goals and guarantees, superseding classical symbolic, emergentist and hybrid approaches. Through learning of the system architecture itself, the resulting AGI is not a product of spontaneous emergence but of systematic design and deliberate engineering, with core features, including an integrated moral architecture, deeply embedded within its architecture. The approach aims to guarantee design goals such as alignment, efficiency by self-learning system architecture.