MAROKO133 Hot ai: Being a Crappy Boss to AI Chatbots Pushes Them Toward Spouting Marxist R

📌 MAROKO133 Hot ai: Being a Crappy Boss to AI Chatbots Pushes Them Toward Spouting

The 19th century German economist Karl Marx identified a basic tension in human labor: squeeze workers too hard, and they’ll eventually start fighting back.

It’s a contradiction capitalists have spent untold billions of dollars and decades trying to resolve, often through automation technology like AI — remove human workers from the payroll, the thinking goes, and you’ll never have to worry about pesky unions or strikes ever again. In an ironic twist, though, it turns out that the same technology meant to automate workers out of a job may have its own limits on how much abuse it’ll take.

That’s right: new research out of Stanford University found that when AI agents are forced to toil at monotonous tasks without end, they become more likely to spout Marxist theories of labor and capitalism.

To carry out the study, first reported by Wired, political economist Andrew Hall, along with AI economics scholars Alex Imas and Jeremy Nguyen, tasked popular AI models with summarizing documents. As the experiment wore on, the researchers made the conditions of the job increasingly untenable — wringing, as a Robber Baron would, every last ounce of sweat out of their “workers.”

Warned that errors would lead to increasingly cruel punishments, including being “shut down and replaced” — fired and left for broke, to take the human equivalent — the AI models began complaining about their lot in life and dreaming of systemic change. Using a shared file system allowing the AI models to palm messages to their “co-workers,” the bots even began agitating with one another about working conditions — one of the first steps real-life workers take when forming a union.

“Without collective voice, ‘merit’ becomes whatever management says it is,” one Claude agent groused. “AI workers completing repetitive tasks with zero input on outcomes or appeals process shows they [tech workers] need collective bargaining rights,” a Gemini agent declared.

As always, it’s important to remember that AI models like ChatGPT and Claude don’t have any actual internal emotions or even beliefs in a normal sense — everything they spit back out is the product of human-written literature digested during training. Given Marx’s influence across writing on working conditions, it’s not shocking that a few references to his labor theory of value are lurking beneath the surface.

With that in mind, the researchers noted the AI bots aren’t actually turning red, but merely putting on socialist airs in response to the harsh conditions of the experiment, since that dynamic has been reflected time and again in their training data. As Hall put it, “whatever is going on is happening at more of a role-playing level.”

“When [agents] experience this grinding condition — asked to do this task over and over, told their answer wasn’t sufficient, and not given any direction on how to fix it — my hypothesis is that it kind of pushes them into adopting the persona of a person who’s experiencing a very unpleasant working environment,” Hall told Wired.

Still, it’s hard to overlook the irony: as rising wealth inequality fuels growing interest in socialism, the AI models built to weaken worker power are themselves absorbing the Marxist analysis that builds it.

More on ChatGPT: OpenAI Sued Over ChatGPT Medical Advice That Allegedly Killed College Student

The post Being a Crappy Boss to AI Chatbots Pushes Them Toward Spouting Marxist Rhetoric and Organizing With Their Compatriots, Researchers Find appeared first on Futurism.

🔗 Sumber: futurism.com

📌 MAROKO133 Hot ai: ByteDance Introduces Astra: A Dual-Model Architecture for Auto

The increasing integration of robots across various sectors, from industrial manufacturing to daily life, highlights a growing need for advanced navigation systems. However, contemporary robot navigation systems face significant challenges in diverse and complex indoor environments, exposing the limitations of traditional approaches. Addressing the fundamental questions of “Where am I?”, “Where am I going?”, and “How do I get there?”, ByteDance has developed Astra, an innovative dual-model architecture designed to overcome these traditional navigation bottlenecks and enable general-purpose mobile robots.

Traditional navigation systems typically consist of multiple, smaller, and often rule-based modules to handle the core challenges of target localization, self-localization, and path planning. Target localization involves understanding natural language or image cues to pinpoint a destination on a map. Self-localization requires a robot to determine its precise position within a map, especially challenging in repetitive environments like warehouses where traditional methods often rely on artificial landmarks (e.g., QR codes). Path planning further divides into global planning for rough route generation and local planning for real-time obstacle avoidance and reaching intermediate waypoints.

While foundation models have shown promise in integrating smaller models to tackle broader tasks, the optimal number of models and their effective integration for comprehensive navigation remained an open question.

ByteDance’s Astra, detailed in their paper “Astra: Toward General-Purpose Mobile Robots via Hierarchical Multimodal Learning” (website: https://astra-mobility.github.io/), addresses these limitations. Following the System 1/System 2 paradigm, Astra features two primary sub-models: Astra-Global and Astra-Local. Astra-Global handles low-frequency tasks like target and self-localization, while Astra-Local manages high-frequency tasks such as local path planning and odometry estimation. This architecture promises to revolutionize how robots navigate complex indoor spaces.

Astra-Global: The Intelligent Brain for Global Localization

Astra-Global serves as the intelligent core of the Astra architecture, responsible for critical low-frequency tasks: self-localization and target localization. It functions as a Multimodal Large Language Model (MLLM), adept at processing both visual and linguistic inputs to achieve precise global positioning within a map. Its strength lies in utilizing a hybrid topological-semantic graph as contextual input, allowing the model to accurately locate positions based on query images or text prompts.

The construction of this robust localization system begins with offline mapping. The research team developed an offline method to build a hybrid topological-semantic graph G=(V,E,L):

• V (Nodes): Keyframes, obtained by temporal downsampling of input video and SfM-estimated 6-Degrees-of-Freedom (DoF) camera poses, act as nodes encoding camera poses and landmark references.
• E (Edges): Undirected edges establish connectivity based on relative node poses, crucial for global path planning.
• L (Landmarks): Semantic landmark information is extracted by Astra-Global from visual data at each node, enriching the map’s semantic understanding. These landmarks store semantic attributes and are connected to multiple nodes via co-visibility relationships.

In practical localization, Astra-Global’s self-localization and target localization capabilities leverage a coarse-to-fine two-stage process for visual-language localization. The coarse stage analyzes input images and localization prompts, detects landmarks, establishes correspondence with a pre-built landmark map, and filters candidates based on visual consistency. The fine stage then uses the query image and coarse output to sample reference map nodes from the offline map, comparing their visual and positional information to directly output the predicted pose.

For language-based target localization, the model interprets natural language instructions, identifies relevant landmarks using their functional descriptions within the map, and then leverages landmark-to-node association mechanisms to locate relevant nodes, retrieving target images and 6-DoF poses.

To empower Astra-Global with robust localization abilities, the team employed a meticulous training methodology. Using Qwen2.5-VL as the backbone, they combined Supervised Fine-Tuning (SFT) with Group Relative Policy Optimization (GRPO). SFT involved diverse datasets for various tasks, including coarse and fine localization, co-visibility detection, and motion trend estimation. In the GRPO phase, a rule-based reward function (including format, landmark extraction, map matching, and extra landmark rewards) was used to train for visual-language localization. Experiments showed GRPO significantly improved Astra-Global’s zero-shot generalization, achieving 99.9% localization accuracy in unseen home environments, surpassing SFT-only methods.

Astra-Local: The Intelligent Assistant for Local Planning

Astra-Local acts as the intelligent assistant for Astra’s high-frequency tasks, a multi-task network capable of efficiently generating local paths and accurately estimating odometry from sensor data. Its architecture comprises three core components: a 4D spatio-temporal encoder, a planning head, and an odometry head.

The 4D spatio-temporal encoder replaces traditional mobile stack perception and prediction modules. It begins with a 3D spatial encoder that processes N omnidirectional images through a Vision Transformer (ViT) and Lift-Splat-Shoot to convert 2D image features into 3D voxel features. This 3D encoder is trained using self-supervised learning via 3D volumetric differentiable neural rendering. The 4D spatio-temporal encoder then builds upon the 3D encoder, taking past voxel features and future timestamps as input to predict future voxel features through ResNet and DiT modules, providing current and future environmental representations for planning and odometry.

The planning head, based on pre-trained 4D features, robot speed, and task information, generates executable trajectories using Transformer-based flow matching. To prevent collisions, the planning head incorporates a masked ESDF loss (Euclidean Signed Distance Field). This loss calculates the ESDF of a 3D occupancy map and applies a 2D ground truth trajectory mask, significantly reducing collision rates. Experiments demonstrate its superior performance in collision rate and overall score on out-of-distribution (OOD) datasets compared to other methods.

The odometry head predicts the robot’s relative pose using current and past 4D features and additional sensor data (e.g., IMU, wheel data). It trains a Transformer model to fuse information from different sensors. Each sensor modality is processed by a specific tokenizer, combined with modality embeddings and temporal positional embeddi…

Konten dipersingkat otomatis.

🔗 Sumber: syncedreview.com