The Machine Intelligence Research Institute (MIRI), formerly the Singularity Institute for Artificial Intelligence (SIAI), is a non-profit research institute focused since 2005 on identifying and managing potential existential risks from artificial general intelligence. MIRI's work has focused on a friendly AI approach to system design and on predicting the rate of technology development.
Verifying claims about AI workloads is a pre- requisite for credible AI governance of covert adversaries (who comply with monitoring only when detection likelihood is high), yet the ap- parent non-determinism of GPU floating-point arithmetic forces auditors to accept approximate output matches. Covert adversaries can exploit un- verifiable degrees of freedom in monitored compu- tation. Attack vectors include steganography, un- reported modification of inference software, and covert computation via unreported batch elements. Empirically, we analyze how modern inference engines (vLLM, HF transformers) produce deter- ministic but non-invariant outputs, without need- ing to set performance-compromising determin- ism flags, if the right information is available for re-computation and no atomic functions are called in the backend. We demonstrate that such bitwise- precise re-computation does not require access to identical hardware, via a software-only emula- tion of LLM inference across multiple NVIDIA GPU variants. Thus, accumulated rounding errors can be an auditable signature of the software and hardware setup used for inference, instead of a constraint on verifiability.
Hardware-enabled monitoring of GPU workloads underpins many proposals for AI compute governance, but if developers can defeat monitoring mechanisms, such schemes are unworkable. We evaluate the adversarial robustness of GPU workload classification using only zero-overhead, privacy-preserving NVML telemetry: content-agnostic signals that observe physical effects of computation without accessing model weights, training data, or hyperparameters. Across 5 rounds of monitor-evader iteration, we evaluate 20 evasion strategy families on 9 GPU models spanning 4 architecture generations. We develop a classifier that achieves 98.2
Appropriate routing strategies are necessary for mobile ad hoc networks (MANETs) in order to facilitate effective data transfer. In order to counter the prevailing problems, the correct routing schemes will need to be selected as the default configurations are used. In this paper, a special optimal link state routing (OLSR) protocol is proposed to incorporate a deep learning methodology to facilitate efficient video streaming in MANETs. This study presents a new improved variant of the OLSR protocol, which is specially tailored to achieve efficient video streaming in MANETs. It is a radical approach that combines a deep-learning model with blockchain technology to overcome security and reliability issues. It starts with the gathering of video content that is available publicly. In order to detect black-hole nodes, a special twin-attention-based Elman spiking neural network model is applied. The reliability of the neighboring nodes is then measured by means of trust values. The pufferfish optimization algorithm, or the accuracy-aware energy-efficient multipath routing algorithm (AEMRAP), which takes into account node- and link-stability degrees, is used in making routing decisions. Interplanetary file system (IPFS) technology is used to store the data on blockchain and increase its security. The authentication of the blockchain architecture is conducted via the delegated proof-of-stake (DPoS) method that also delivers an extra protection of MANETs against unauthorized access. The study demonstrates superior performance in securing and optimizing video transmission, confirming that the extended OLSR protocol is highly effective for MANET video streaming applications. The proposed model exceeds the current approaches with a throughput of 2100 Kbps, an average end latency of 20.2 s, and a packet-delivery ratio of 92.3%.
The premature development of artificial superintelligence poses major risks to humanity, so researchers have proposed international agreements halting such development until it can be done safely. AI progress depends primarily on compute, algorithms, and data; a durable halt would address all three so that advances in one input do not counteract restrictions on another. Improvements to AI algorithms are driven largely through research activities, so this research may need to be restricted during a halt. Given low international trust, signatories will want to verify compliance. This paper analyzes how such restrictions on AI research could be verified, while remaining agnostic about what specific research would be prohibited. It first explores key considerations that affect the verifiability of research restrictions, such as the computational infrastructure necessary for experiments. It then catalogs 28 candidate verification mechanisms. These mechanisms include whistleblowers, search warrants, reviews of AI training code, standard intelligence gathering tools, and more. Some of these mechanisms are not yet implementation-ready, and some might be undesirable upon further inspection. By examining the space of potential options, this work provides a foundation for future research to develop the most promising mechanisms into deployable tools.
As AI capabilities advance, AI systems will pose greater risks to national security and potentially humanity as a whole. Governments may eventually conclude that these risks warrant restraining AI development. This motivates the question: will governments still be able to restrain AI development in the future, should they want to do so? In this paper we analyze which world events and changes to the state of AI development would make future governance more difficult or even effectively impossible. Our analysis surfaces likely pathways that would lead to these difficulties, including hardware proliferation, continued algorithmic progress, and the release of catastrophically dangerous AI models. Due to the field's lack of understanding of AI development, it may be difficult or impossible to know when we will hit a "point of no return", and we therefore recommend a conservative approach. The window may be closing, but governments currently have an opportunity to preserve their optionality if they act soon. Our policy recommendations would enable governments to restrain AI development in the future, while imposing relatively small costs today.
