Lifeboat Foundation: Safeguarding Humanity
Toggle light / dark theme

Blog

Category: robotics/AI – Page 75

Microsoft: SesameOp malware abuses OpenAI Assistants API in attacks

Microsoft security researchers have discovered a new backdoor malware that uses the OpenAI Assistants API as a covert command-and-control channel.

The company’s Detection and Response Team (DART) discovered the new malware, named SesameOp, during an investigation into a July 2025 cyberattack, which revealed that the malware allowed attackers to gain persistent access to the compromised environment.

Deploying this malware also enabled the threat actors to remotely manage backdoored devices for several months by leveraging legitimate cloud services, rather than relying on dedicated malicious infrastructure that could alert victims to an attack and be taken down during subsequent incident response.

New Artificial Neurons Physically Replicate the Brain

A breakthrough in neuromorphic computing could lower the energy consumption of chips and accelerate progress toward artificial general intelligence (AGI). Researchers from the USC Viterbi School of Engineering and the School of Advanced Computing have created artificial neurons that closely mimic

Student trust in AI coding tools grows briefly, then levels off with experience

How much do undergraduate computer science students trust chatbots powered by large language models like GitHub Copilot and ChatGPT? And how should computer science educators modify their teaching based on these levels of trust?

These were the questions that a group of U.S. computer scientists set out to answer in a study that will be presented at the Koli Calling conference Nov. 11 to 16 in Finland. In the course of the study’s few weeks, researchers found that trust in generative AI tools increased in the short run for a majority of students.

But in the long run, students said they realized they needed to be competent programmers without the help of AI tools. This is because these tools often generate incorrect or would not help students with code comprehension tasks.

Fintech’s New Power Couple: AI And Trust

At Money20/20 I learned that Fintech has a new power couple, AI and Trust.

The combination of the two is the payment protocol of tomorrow. Come along with me as I share my findings from the world’s #1 fintech show.

There is bunch of interviews and images forthcoming from the event.

Thanks again to Tedd Huff of Fintech Confidential for inviting me to participate in the event. It allowed me to share My Instant AI with event attendees.

(https://www.linkedin.com/pulse/fintechs-new-power-couple-ai-…urke-eirte)

Tedd Huff asked me to be a confidential informant in Las Vegas recently. But, wait before you go down conspiracy theory rabbit hole, please let me explain.

Continue reading “Fintech’s New Power Couple: AI And Trust” | >

Topographical sparse mapping: A neuro-inspired sparse training framework for deep learning models

AI models have been expanding dramatically in size and the number of trainable parameters. This rapid growth has introduced many challenges, including increased computational costs and inefficiencies. Dynamic sparse training has emerged as a novel approach to address overparameterization and achieve energy-efficient artificial neural network (ANN) architectures. The highly efficient neuro-inspired sparse design remains underexplored compared to the significant focus on random topology searches. We propose the Topographical Sparse Mapping (TSM) method, inspired by the vertebrate visual system and convergent units. TSM introduces a sparse input layer for MLPs, significantly reducing the number of parameters.

Artificial muscles use ultrasound-activated microbubbles to move

Researchers at ETH Zurich have developed artificial muscles that contain microbubbles and can be controlled with ultrasound. In the future, these muscles could be deployed in technical and medical settings as gripper arms, tissue patches, targeted drug delivery, or robots.

It might look like a simple material experiment at first glance, as a brief ultrasound stimulation induces a thin strip of silicone to start bending and arching. But that’s just the beginning. A team led by Daniel Ahmed, Professor of Acoustic Robotics for Life Sciences and Healthcare, has developed a new class of : flexible membranes that respond to ultrasound with the help of thousands of microbubbles.

The work is published in the journal Nature.

AI model identifies high-performing battery electrolytes by starting from just 58 data points

In an ideal world, an AI model looking for new materials to build better batteries would be trained on millions or even hundreds of millions of data points.

But for emerging next-generation battery chemistries that don’t have decades of research behind them, waiting for new studies takes time the world doesn’t have.

“Each experiment takes up to weeks, months to get ,” said University of Chicago Pritzker School of Molecular Engineering (UChicago PME) Schmidt AI in Science Postdoctoral Fellow Ritesh Kumar. “It’s just infeasible to wait until we have millions of data to train these models.”

/* */