Post Syndicated from Bruce Schneier original https://www.schneier.com/blog/archives/2025/05/florida-backdoor-bill-fails.html
A Florida bill requiring encryption backdoors failed to pass.
Tag Archives: Uncategorized
Friday Squid Blogging: Japanese Divers Video Giant Squid
Post Syndicated from Bruce Schneier original https://www.schneier.com/blog/archives/2025/05/friday-squid-blogging-japanese-divers-video-giant-squid.html
The video is really amazing.
As usual, you can also use this squid post to talk about the security stories in the news that I haven’t covered.
Chinese AI Submersible
Post Syndicated from Bruce Schneier original https://www.schneier.com/blog/archives/2025/05/chinese-ai-submersible.html
A Chinese company has developed an AI-piloted submersible that can reach speeds “similar to a destroyer or a US Navy torpedo,” dive “up to 60 metres underwater,” and “remain static for more than a month, like the stealth capabilities of a nuclear submarine.” In case you’re worried about the military applications of this, you can relax because the company says that the submersible is “designated for civilian use” and can “launch research rockets.”
“Research rockets.” Sure.
Fake Student Fraud in Community Colleges
Post Syndicated from Bruce Schneier original https://www.schneier.com/blog/archives/2025/05/fake-student-fraud-in-community-colleges.html
Reporting on the rise of fake students enrolling in community college courses:
The bots’ goal is to bilk state and federal financial aid money by enrolling in classes, and remaining enrolled in them, long enough for aid disbursements to go out. They often accomplish this by submitting AI-generated work. And because community colleges accept all applicants, they’ve been almost exclusively impacted by the fraud.
The article talks about the rise of this type of fraud, the difficulty of detecting it, and how it upends quite a bit of the class structure and learning community.
Slashdot thread.
Another Move in the Deepfake Creation/Detection Arms Race
Post Syndicated from Bruce Schneier original https://www.schneier.com/blog/archives/2025/05/another-move-in-the-deepfake-creation-detection-arms-race.html
Deepfakes are now mimicking heartbeats
In a nutshell
- Recent research reveals that high-quality deepfakes unintentionally retain the heartbeat patterns from their source videos, undermining traditional detection methods that relied on detecting subtle skin color changes linked to heartbeats.
- The assumption that deepfakes lack physiological signals, such as heart rate, is no longer valid. This challenges many existing detection tools, which may need significant redesigns to keep up with the evolving technology.
- To effectively identify high-quality deepfakes, researchers suggest shifting focus from just detecting heart rate signals to analyzing how blood flow is distributed across different facial regions, providing a more accurate detection strategy.
And the AI models will start mimicking that.
Friday Squid Blogging: Pyjama Squid
Post Syndicated from Bruce Schneier original https://www.schneier.com/blog/archives/2025/05/friday-squid-blogging-pyjama-squid.html
The small pyjama squid (Sepioloidea lineolata) produces toxic slime, “a rare example of a poisonous predatory mollusc.”
As usual, you can also use this squid post to talk about the security stories in the news that I haven’t covered.
Privacy for Agentic AI
Post Syndicated from Bruce Schneier original https://www.schneier.com/blog/archives/2025/05/privacy-for-agentic-ai.html
Sooner or later, it’s going to happen. AI systems will start acting as agents, doing things on our behalf with some degree of autonomy. I think it’s worth thinking about the security of that now, while its still a nascent idea.
In 2019, I joined Inrupt, a company that is commercializing Tim Berners-Lee’s open protocol for distributed data ownership. We are working on a digital wallet that can make use of AI in this way. (We used to call it an “active wallet.” Now we’re calling it an “agentic wallet.”)
I talked about this a bit at the RSA Conference earlier this week, in my keynote talk about AI and trust. Any useful AI assistant is going to require a level of access—and therefore trust—that rivals what we currently our email provider, social network, or smartphone.
This Active Wallet is an example of an AI assistant. It’ll combine personal information about you, transactional data that you are a party to, and general information about the world. And use that to answer questions, make predictions, and ultimately act on your behalf. We have demos of this running right now. At least in its early stages. Making it work is going require an extraordinary amount of trust in the system. This requires integrity. Which is why we’re building protections in from the beginning.
Visa is also thinking about this. It just announced a protocol that uses AI to help people make purchasing decisions.
I like Visa’s approach because it’s an AI-agnostic standard. I worry a lot about lock-in and monopolization of this space, so anything that lets people easily switch between AI models is good. And I like that Visa is working with Inrupt so that the data is decentralized as well. Here’s our announcement about its announcement:
This isn’t a new relationship—we’ve been working together for over two years. We’ve conducted a successful POC and now we’re standing up a sandbox inside Visa so merchants, financial institutions and LLM providers can test our Agentic Wallets alongside the rest of Visa’s suite of Intelligent Commerce APIs.
For that matter, we welcome any other company that wants to engage in the world of personal, consented Agentic Commerce to come work with us as well.
I joined Inrupt years ago because I thought that Solid could do for personal data what HTML did for published information. I liked that the protocol was an open standard, and that it distributed data instead of centralizing it. AI agents need decentralized data. “Wallet” is a good metaphor for personal data stores. I’m hoping this is another step towards adoption.
NCSC Guidance on “Advanced Cryptography”
Post Syndicated from Bruce Schneier original https://www.schneier.com/blog/archives/2025/05/ncsc-guidance-on-advanced-cryptography.html
The UK’s National Cyber Security Centre just released its white paper on “Advanced Cryptography,” which it defines as “cryptographic techniques for processing encrypted data, providing enhanced functionality over and above that provided by traditional cryptography.” It includes things like homomorphic encryption, attribute-based encryption, zero-knowledge proofs, and secure multiparty computation.
It’s full of good advice. I especially appreciate this warning:
When deciding whether to use Advanced Cryptography, start with a clear articulation of the problem, and use that to guide the development of an appropriate solution. That is, you should not start with an Advanced Cryptography technique, and then attempt to fit the functionality it provides to the problem.
And:
In almost all cases, it is bad practice for users to design and/or implement their own cryptography; this applies to Advanced Cryptography even more than traditional cryptography because of the complexity of the algorithms. It also applies to writing your own application based on a cryptographic library that implements the Advanced Cryptography primitive operations, because subtle flaws in how they are used can lead to serious security weaknesses.
The conclusion:
Advanced Cryptography covers a range of techniques for protecting sensitive data at rest, in transit and in use. These techniques enable novel applications with different trust relationships between the parties, as compared to traditional cryptographic methods for encryption and authentication.
However, there are a number of factors to consider before deploying a solution based on Advanced Cryptography, including the relative immaturity of the techniques and their implementations, significant computational burdens and slow response times, and the risk of opening up additional cyber attack vectors.
There are initiatives underway to standardise some forms of Advanced Cryptography, and the efficiency of implementations is continually improving. While many data processing problems can be solved with traditional cryptography (which will usually lead to a simpler, lower-cost and more mature solution) for those that cannot, Advanced Cryptography techniques could in the future enable innovative ways of deriving benefit from large shared datasets, without compromising individuals’ privacy.
NCSC blog entry.
US as a Surveillance State
Post Syndicated from Bruce Schneier original https://www.schneier.com/blog/archives/2025/05/us-as-a-surveillance-state.html
Two essays were just published on DOGE’s data collection and aggregation, and how it ends with a modern surveillance state.
It’s good to see this finally being talked about.
EDITED TO ADD (5/3): Here’s a free link to that first essay.
AWS’s Well-Architected Framework Transformed by Amazon Q Developer
Post Syndicated from Fahim Sajjad original https://aws.amazon.com/blogs/devops/awss-well-architected-framework-transformed-by-amazon-q-developer/
In the rapidly evolving landscape of cloud computing, developers, and architects face unprecedented challenges. These challenges include designing, implementing, and maintaining robust cloud infrastructures. The AWS Well-Architected Framework is the gold standard for building secure, efficient, and optimized cloud solutions. Traditionally, complying with this framework required deep expertise and manual analysis.
Now, Amazon Q Developer changes this paradigm. It introduces intelligent, context-aware recommendations. This framework is built on six pillars: operational excellence, security, reliability, performance efficiency, cost optimization, and sustainability. Without careful attention to these foundational elements, organizations risk developing systems that fall short of both their business objectives and technical requirements, potentially compromising long-term scalability and performance.
In this blog, we will explore how Amazon Q Developer can help across the six pillars of the Well-Architected Framework.
Simplifying the AWS Well-Architected Framework with Amazon Q Developer
In this section, we will explore practical examples across the following pillars:
- Operational excellence pillar with Amazon Q Developer
- Security pillar with Amazon Q Developer
- Reliability pillar with Amazon Q Developer
- Performance efficiency pillar with Amazon Q Developer
- Cost optimization pillar with Amazon Q Developer
- Sustainability pillar with Amazon Q Developer
Operational Excellence pillar with Amazon Q Developer
Operational Excellence is a key pillar of the AWS Well-Architected Framework. It guides teams in running efficient workloads and monitoring systems. The pillar emphasizes continuous improvement to deliver business value. Amazon Q Developer enhances operations through AI-powered assistance for infrastructure as code, automated testing, and documentation. The service automatically creates runbooks and suggests safe infrastructure changes. It analyzes your environment and offers recommendations based on AWS best practices. Teams can implement operational excellence with reduced manual effort and fewer potential errors.
Let’s ask Amazon Q Developer in the console how it can help me improve operational excellence in my AWS Infrastructure: “How can I improve the Operational Excellence of my AWS infrastructure?”
Amazon Q Developer analyzes your prompt and generates comprehensive recommendations:
Figure 1: Prompting Amazon Q about Operational Excellence
Based on Amazon Q Developer’s guidance, Infrastructure as Code (IaC) is recommended for managing our AWS resources. Let’s open Visual Studio Code IDE and see how Amazon Q Developer Chat can help us implement IaC. We’ll create a CloudFormation template for a resilient web application by asking Amazon Q Developer to generate a template that includes an Application Load Balancer, EC2 instances, and an RDS database: “Generate a CloudFormation template for a highly available web application with an Application Load Balancer, EC2 instances, and an RDS database. Include best practices for operational excellence”
Figure 2: Amazon Q Developer generating a CloudFormation template for a highly available web application architecture.
It also explains the template and lists the best practices for operational excellence that were followed:
Figure 3: Amazon Q explaining Generated template and Best Practices
Amazon Q Developer automates documentation to track and approve infrastructure changes effectively. It performs automatic code reviews to check quality and identify security vulnerabilities. The service detects code duplication and guides developers in making small, predictable changes. When issues arise, Amazon Q Developer quickly investigates operational problems across AWS environments. This rapid troubleshooting helps businesses maintain high application availability.
Security pillar with Amazon Q Developer
Cloud security at AWS is the highest priority. The AWS Well-Architected Framework’s Security Pillar provides a comprehensive approach to implementing robust protective measures. Even though traditionally security has been an afterthought in development often sacrificed for speed and automation, Amazon Q Developer transforms this by enabling security checks at every stage of the software development lifecycle. By embedding continuous security validation, you can significantly reduce vulnerabilities in production environments. This shifts security from reactive to proactive, ensuring your cloud applications are not just functional, but fundamentally secure.
Amazon Q Developer can streamline this process serving as an intelligent security assistant for AWS environments. It can help create robust Identity and Access Management (IAM) policies, including role-based access controls, user permissions, and multi-factor authentication. For data protection, Amazon Q Developer supports encryption strategies, key management, and secure backup procedures. Additionally, Amazon Q Developer guides users in infrastructure protection planning through secure network architectures and VPC segmentation, offering comprehensive support across critical security areas.
Amazon Q Developer enhances security beyond basic configurations. It helps set up advanced monitoring solutions using CloudWatch and CloudTrail. The service creates intelligent security alerts and automates incident response mechanisms. It protects your AWS environment against emerging threats through security scanning. Amazon Q Developer identifies potential vulnerabilities in your infrastructure. These capabilities align with AWS Well-Architected Framework security best practices. To illustrate Amazon Q Developer’s practical application in enhancing workload security, let’s consider implementing VPC flow logs for improved network monitoring in our web application CloudFormation file that we created before: “How can we implement VPC flow logs for better network monitoring?”
Figure 4: Prompting Amazon Q about implementing VPC flow logs for better network monitoring
We can also ask Amazon Q Developer to check if there are any security best practices that is missing in our code. “What IAM security best practices are missing?”
Figure 5: Prompting Amazon Q about implementing VPC flow logs for better network monitoring.
These examples demonstrates how developers can leverage Amazon Q Developer to bolster their security posture effectively.
Reliability pillar with Amazon Q Developer
Reliability is a critical pillar of the AWS Well-Architected Framework. It extends beyond maintaining system uptime. A reliable architecture must handle unexpected disruptions and recover gracefully. Amazon Q Developer brings AI-powered intelligence to reliability engineering. It helps organizations design resilient, self-healing cloud infrastructures. The service anticipates potential failures and suggests mitigation strategies before business operations are affected.
Amazon Q Developer guides users in implementing AWS reliability best practices. It helps architect resilient systems through multi-AZ deployments and auto-scaling configurations. The service assists in setting up automated recovery procedures and fault-tolerant systems. It streamlines the configuration of health checks, load balancers, and redundant components. Amazon Q Developer enables effective monitoring through CloudWatch alarms and automated failover mechanisms. It supports infrastructure as code implementation with proper testing procedures. The service helps establish cross-region redundancy and appropriate service quotas. These features ensure systems maintain high availability and recover quickly from failures.
By interacting with Amazon Q Developer in the AWS Management Console, you can receive intelligent recommendations for improving your infrastructure’s reliability by asking Amazon Q Developer: “Can you provide recommendations to eliminate single point of failures?”
Figure 6: Prompting Amazon Q about reliabilityPerformance Efficiency pillar with Amazon Q Developer
Performance can determine an application’s success in cloud computing. The Performance Efficiency pillar guides organizations in maximizing their computational resources. Amazon Q Developer enhances this pillar through AI-powered recommendations. It helps organizations design and optimize their cloud infrastructure more effectively.
Amazon Q Developer uses machine learning to deliver advanced performance insights. It recommends architectural improvements like serverless adoption and optimal service configurations. The AI assistant suggests effective caching strategies and data processing optimizations. It analyzes system metrics and infrastructure patterns to guide improvements. Development teams can enhance system performance and reduce computational overhead. Amazon Q Developer helps create adaptive architectures that respond effectively to changing workload demands.
For example, let’s say you’re an IT Professional with a monolithic three-tier web application on AWS and wanted to get recommendation on performance efficiency. The IT Professional could open a new Amazon Q Developer chat in the AWS Management Console, and enter a prompt such as: “Based on my current monolithic application on AWS, what are some things I should do as it relates to the performance efficiency pillar of the Well-Architected Framework?”
Figure 7: Prompting Amazon Q about performance efficiency
As shown above in figure 7, Amazon Q Developer made multiple recommendations based on the monolith application in the prompt. Amazon Q Developer makes recommendations such as using Serverless options and breaking the application into microservices, which is a design principal in the performance efficiency pillar of the Well-Architected Framework.
Cost Optimization pillar with Amazon Q Developer
Amazon Q Developer has the ability to give general recommendations for Cost Optimization based on the Well-Architected Framework. For example, let’s say an IT Professional wants to get more information about ways they can generally optimize their compute costs in AWS. The IT Professional could open a new Amazon Q chat in the AWS Management Console, and enter a prompt such as: “What are some ways I can cost optimize my compute infrastructure I have running in AWS based on the cost optimization pillar of the Well Architected Framework?”
Figure 8: Prompting Amazon Q about cost optimization
As shown above in figure 8, Amazon Q Developer was able to make recommendations based on the Well-architected framework to help the developer cost optimize their compute through reserved instances and savings plans, auto scaling, rightsizing, and more, while also providing links to resources to help dive in further.
Additionally, Amazon Q Developer has revolutionized AWS cost analysis by introducing natural language processing capabilities directly integrated with AWS Cost Explorer. This innovative feature allows users to gain deep insights into cloud spending through simple, conversational queries, enabling professionals to understand complex cost structures, identify spending trends, and forecast future expenses with unprecedented ease. By transforming technical cost data into actionable insights, Amazon Q Developer empowers organizations to make more informed financial decisions about their cloud infrastructure.
For comprehensive details and specific use case, please refer to the full blog post: Analyzing your AWS Cost Explorer data with Amazon Q Developer: Now Generally Available.
Sustainability pillar with Amazon Q Developer
Sustainability represents a critical emerging pillar of the AWS Well-Architected Framework, focusing on minimizing the environmental impact of cloud computing infrastructure and operations. Amazon Q Developer introduces AI-powered capabilities that help organizations optimize their cloud resources to reduce carbon footprint, improve energy efficiency, and align technological strategies with environmental responsibility.
Through intelligent analysis and context-aware recommendations, Amazon Q Developer enables teams to make more sustainable architectural decisions. The AI assistant can provide insights into resource optimization, suggesting ways to reduce unnecessary compute power, recommend more energy-efficient service configurations, and help developers understand the environmental implications of their architectural choices. By leveraging machine learning and comprehensive AWS infrastructure knowledge, Amazon Q Developer empowers organizations to not only meet their technological requirements but also contribute to broader environmental sustainability goals in cloud computing.
In the below example, you can see how you can ask Amazon Q Developer to help meet your company’s sustainability goals by asking Amazon Q: “How can I review my sustainability objectives on AWS?”
Figure 9: Prompting Amazon Q about sustainability
As you can see above Amazon Q Developer generated recommendations showing how you can review your sustainability objectives.
Now let’s take the first recommendation: Carbon Footprint tool as an example and ask Amazon Q Developer in the console to ask a follow up question We will be using the following prompt to generate response: “How do I view my carbon footprint on AWS?”
Figure 10: Prompting Amazon Q about carbon footprint
Conclusion
Amazon Q Developer represents a pivotal moment in cloud computing, transforming the AWS Well-Architected Framework from a static set of guidelines to a dynamic, intelligent system of continuous improvement. By integrating advanced AI capabilities across operational excellence, security, reliability, performance efficiency, cost optimization, and sustainability, this innovative tool democratizes sophisticated cloud architecture strategies for organizations of all sizes. The true power of Amazon Q Developer lies not just in its ability to provide recommendations, but in its capacity to learn, adapt, and evolve alongside your infrastructure, bridging the gap between complex technical knowledge and actionable insights. As cloud technologies continue to advance, AI-powered tools like Amazon Q Developer will become increasingly essential, signifying a fundamental shift in how we approach cloud infrastructure: proactively, intelligently, and with a holistic understanding of technological and business requirements.
To get started with Amazon Q Developer in the AWS console, check out the documentation on chatting with Amazon Q Developer in AWS Console Home.
Migrating a CDK v1 Application to CDK v2 with Amazon Q Developer
Post Syndicated from Dr. Rahul Sharad Gaikwad original https://aws.amazon.com/blogs/devops/migrating-a-cdk-v1-application-to-cdk-v2-with-amazon-q-developer/
Introduction:
AWS Cloud Development Kit (AWS CDK) is an open-source software development framework for defining cloud infrastructure in code and provisioning it through AWS CloudFormation. As of June 1, 2023, AWS CDK version 1 is no longer supported. To avoid the potential issues that come with using an outdated version and to take advantage of the latest features and improvements, we highly recommend upgrading to AWS CDK version 2.
Amazon Q Developer, a generative AI-powered assistant for software development, enhances the efficiency of software development teams. It facilitates the creation of deployment-ready infrastructure as code (IaC) for AWS CloudFormation, AWS CDK, and Terraform. By using Amazon Q, developers can accelerate IaC development, enhance code quality, and decrease the likelihood of configuration errors.
This post demonstrates how Amazon Q Developer helps in upgrading the existing AWS CDK v1 application to AWS CDK v2.
Prerequisites
- An AWS Builder ID or an AWS IAM Identity Center login controlled by your organization
- A supported IDE, like Visual Studio Code
- The AWS Toolkit IDE extension
- Authenticate and Connect
- Nodejs
- AWS CDK v1
- AWS CDK v2
Planning
In this blog post, I will explore a code example where I have created a VPC, Subnets, and an ECS Fargate cluster using AWS CDK version 1. I will then explain how you can use Amazon Q to transform the code from CDK v1 to CDK v2.
1. In order to initiate this process, I have begun by asking Amazon Q Developer for the necessary steps to migrate from CDK version 1 to version 2, which are outlined below.
Can you provide the steps to migrate from cdk version 1 to version 2?
2. In the above screenshot Amazon Q Developer outlined several steps we can take to make the necessary changes. The first step is to update the dependencies. If I need guidance on how to update the dependencies, I can ask the Amazon Q Developer again for help by asking the steps regarding updating dependencies as below .
Can you provide the steps to update dependencies?
3. After updating the dependencies, the next step is to update the import statements. To get guidance on how to update the import statements, I can ask the Amazon Q Developer assistant again for help by asking the steps regarding how to import statements as shown below.
@workspace Can you provide the steps to update import statements?
In the above screenshot if you have noticed I have added @workspace before the question which automatically includes the most relevant chunks of my workspace code as context.
4. If any errors occur while updating the code as recommended by Amazon Q Developer, I can use Amazon Q Developer to debug the issue and provide the needed inputs to resolve it.
5. Once I have finished the required steps, I can deploy the application using version 2 of the AWS CDK by running the cdk deploy command.
6. In addition to its other capabilities, Amazon Q offers code review functionality. To initiate a code review, simply select Amazon Q and use the /review command. I’ll then have the option to review either the active files or the entire open workspace. Select your preference, and Amazon Q will analyze your project and provide comprehensive review results.
7. Amazon Q Developer can also generate documentation, including README files. To create documentation, select Amazon Q and enter the /doc command. Amazon Q will automatically generate a README file for your project. I can then review the generated documentation, accept the changes, or provide specific instructions for further modifications.
Conclusion
In this blog, I demonstrated how Amazon Q Developer can simplify and accelerate the upgrade process from AWS CDK version 1 to version 2, ensuring your cloud infrastructure remains secure, efficient, and aligned with the latest AWS innovations. AWS CDK v2 offers a streamlined, consolidated library with improved performance and ongoing support, making infrastructure management easier and more reliable.
By leveraging Amazon Q Developer, a generative AI-powered assistant, teams can automate Infrastructure as Code development, enhance code quality, and minimize configuration errors. Together, these tools empower development teams to confidently modernize and scale their AWS environments, turning the upgrade process into a seamless opportunity for innovation and growth.
Resources
To learn more about Amazon Q Developer, see the following resources:
To learn more about the AWS CDK, see the following resources:
About the authors:
WhatsApp Case Against NSO Group Progressing
Post Syndicated from Bruce Schneier original https://www.schneier.com/blog/archives/2025/04/whatsapp-case-against-nso-group-progressing.html
Meta is suing NSO Group, basically claiming that the latter hacks WhatsApp and not just WhatsApp users. We have a procedural ruling:
Under the order, NSO Group is prohibited from presenting evidence about its customers’ identities, implying the targeted WhatsApp users are suspected or actual criminals, or alleging that WhatsApp had insufficient security protections.
[…]
In making her ruling, Northern District of California Judge Phyllis Hamilton said NSO Group undercut its arguments to use evidence about its customers with contradictory statements.
“Defendants cannot claim, on the one hand, that its intent is to help its clients fight terrorism and child exploitation, and on the other hand say that it has nothing to do with what its client does with the technology, other than advice and support,” she wrote. “Additionally, there is no evidence as to the specific kinds of crimes or security threats that its clients actually investigate and none with respect to the attacks at issue.”
I have written about the issues at play in this case.
Applying Security Engineering to Prompt Injection Security
Post Syndicated from Bruce Schneier original https://www.schneier.com/blog/archives/2025/04/applying-security-engineering-to-prompt-injection-security.html
This seems like an important advance in LLM security against prompt injection:
Google DeepMind has unveiled CaMeL (CApabilities for MachinE Learning), a new approach to stopping prompt-injection attacks that abandons the failed strategy of having AI models police themselves. Instead, CaMeL treats language models as fundamentally untrusted components within a secure software framework, creating clear boundaries between user commands and potentially malicious content.
[…]
To understand CaMeL, you need to understand that prompt injections happen when AI systems can’t distinguish between legitimate user commands and malicious instructions hidden in content they’re processing.
[…]
While CaMeL does use multiple AI models (a privileged LLM and a quarantined LLM), what makes it innovative isn’t reducing the number of models but fundamentally changing the security architecture. Rather than expecting AI to detect attacks, CaMeL implements established security engineering principles like capability-based access control and data flow tracking to create boundaries that remain effective even if an AI component is compromised.
Research paper. Good analysis by Simon Willison.
I wrote about the problem of LLMs intermingling the data and control paths here.
Windscribe Acquitted on Charges of Not Collecting Users’ Data
Post Syndicated from Bruce Schneier original https://www.schneier.com/blog/archives/2025/04/windscribe-acquitted-on-charges-of-not-collecting-users-data.html
The company doesn’t keep logs, so couldn’t turn over data:
Windscribe, a globally used privacy-first VPN service, announced today that its founder, Yegor Sak, has been fully acquitted by a court in Athens, Greece, following a two-year legal battle in which Sak was personally charged in connection with an alleged internet offence by an unknown user of the service.
The case centred around a Windscribe-owned server in Finland that was allegedly used to breach a system in Greece. Greek authorities, in cooperation with INTERPOL, traced the IP address to Windscribe’s infrastructure and, unlike standard international procedures, proceeded to initiate criminal proceedings against Sak himself, rather than pursuing information through standard corporate channels.
Friday Squid Blogging: Squid Facts on Your Phone
Post Syndicated from Bruce Schneier original https://www.schneier.com/blog/archives/2025/04/friday-squid-blogging-squid-facts-on-your-phone.html
Text “SQUID” to 1-833-SCI-TEXT for daily squid facts. The website has merch.
As usual, you can also use this squid post to talk about the security stories in the news that I haven’t covered.
Cryptocurrency Thefts Get Physical
Post Syndicated from Bruce Schneier original https://www.schneier.com/blog/archives/2025/04/cryptocurrency-thefts-get-physical.html
Long story of a $250 million cryptocurrency theft that, in a complicated chain events, resulted in a pretty brutal kidnapping.
New Linux Rootkit
Post Syndicated from Bruce Schneier original https://www.schneier.com/blog/archives/2025/04/new-linux-rootkit.html
The company has released a working rootkit called “Curing” that uses io_uring, a feature built into the Linux kernel, to stealthily perform malicious activities without being caught by many of the detection solutions currently on the market.
At the heart of the issue is the heavy reliance on monitoring system calls, which has become the go-to method for many cybersecurity vendors. The problem? Attackers can completely sidestep these monitored calls by leaning on io_uring instead. This clever method could let bad actors quietly make network connections or tamper with files without triggering the usual alarms.
Here’s the code.
Note the self-serving nature of this announcement: ARMO, the company that released the research and code, has a product that it claims blocks this kind of attack.
Amazon introduces SWE-PolyBench, a multilingual benchmark for AI Coding Agents
Post Syndicated from Christian Bock original https://aws.amazon.com/blogs/devops/amazon-introduces-swe-polybench-a-multi-lingual-benchmark-for-ai-coding-agents/
Coding agents powered by large language models have shown impressive capabilities in software engineering tasks, but evaluating their performance across diverse programming languages and real-world scenarios remains challenging. This led to a recent explosion in benchmark creation to assess the coding effectiveness of said systems in controlled environments. In particular, SWE-Bench which measures the performance of systems in the context of GitHub issues has spurred the development of capable coding agents resulting in over 50 leaderboard submissions, thereby becoming the de-facto standard for coding agent benchmarking. Despite its significant impact as a pioneering benchmark, SWE-Bench, and in particular its “verified” subset, also shows some limitations. It contains only Python repositories, the majority of tasks are bug fixes, and at over 45% of all tasks, the Django repository is significantly over-represented.
Today, Amazon introduces SWE-PolyBench, the first industry benchmark to evaluate AI coding agents’ ability to navigate and understand complex codebases, introducing rich metrics to advance AI performance in real-world scenarios. SWE-PolyBench contains over 2,000 curated issues in four languages. In addition, it contains a stratified subset of 500 issues (SWE-PolyBench500) for the purpose of rapid experimentation. SWE-PolyBench evaluates the performance of AI coding agents through a comprehensive set of metrics: pass rates across different programming languages and task complexity levels, along with precision and recall measurements for code/file context identification. These evaluation metrics can help the community address challenges in understanding how well AI coding agents can navigate through and comprehend complex codebases
The leaderboard is accessible here. The SWE-PolyBench dataset is available on Hugging Face and the paper at arxiv. Evaluations can be run using the SWE-PolyBench codebase.
Below, we describe the key features, characteristics, and the creation process of our dataset alongside the new evaluation metrics, and performance of open source agents from our experiments.
Key features of SWE-PolyBench at a glance
- Multi-Language Support: Java (165 tasks), JavaScript (1017 tasks), TypeScript (729 tasks), and Python (199 tasks).
- Extensive Dataset: 2110 instances from 21 repositories ranging from web frameworks to code editors and ML tools, on the same scale as SWE-Bench full with more repository.
- Task Variety: Includes bug fixes, feature requests, and code refactoring.
- Faster Experimentation: SWE-PolyBench500 is a stratified subset for efficient experimentation.
- Leaderboard: A leaderboard with a rich set of metrics for transparent benchmarking.
Building a comprehensive dataset
The creation of SWE-PolyBench involved a data collection and filtering process designed to ensure the quality and relevance of the benchmark tasks. SWE-Bench, a benchmark for Python code generation, evaluates agents on real-world programming tasks by utilizing GitHub issues and their corresponding code and test modifications. We extended the SWE-Bench data acquisition pipeline to support 3 additional languages besides Python and used it to gather and process coding challenges from real-world repositories as shown in Figure 1.
Figure 1: Overview of the SWE-PolyBench data generation pipeline, illustrating the process of collecting, filtering, and validating coding tasks.
The data acquisition pipeline collects pull requests (PRs) that close issues from popular repositories across Java, JavaScript, TypeScript, and Python. These PRs undergo filtering and are set up in containerized environments for consistent test execution. The process categorizes tests as fail-to-pass (F2P) or pass-to-pass (P2P) based on their outcomes before and after patch application. Only PRs with at least one F2P test are included in the final dataset, ensuring that each task represents a meaningful coding challenge. This streamlined approach results in a dataset that closely mimics real-world coding scenarios, providing a robust foundation for evaluating AI coding assistants.
Dataset characteristics
When constructing SWE-PolyBench, we aimed to collect GitHub issues that represent diverse programming scenarios: issues involving modifications across multiple code files and spanning different task categories (such as bug fixes, feature requests, and refactoring). Tables 1 and 2 provide descriptive statistics on the composition and complexity of SWE-PolyBench full (PB) and SWE-PolyBench500 (PB500). To offer a point of reference, we compare these statistics with those of SWE-Bench (SWE) and SWE-Bench verified (SWEv). Tasks in SWE-PolyBench require on average more files to be modified and more nodes to be changed, which indicates that they have higher complexity and are closer to tasks in real-world projects. The distribution of tasks is also more diverse, in particular for SWE-PolyBench500.
New evaluation metrics
To comprehensively evaluate AI coding assistants, SWE-Polybench introduces multiple new metrics in addition to the pass rate. The pass rate is the proportion of tasks successfully solved as measured by the generated patch passing all relevant tests. It is the primary metric for assessing coding agent performance, but it doesn’t provide a complete picture of an agent’s capabilities. In particular, it doesn’t give much information on an agent’s ability to navigate and understand complex code repositories. SWE-PolyBench introduces a new set of metrics based on Concrete Syntax Tree (CST) node analysis and the established file-level localization metric:
- File-level Localization: assesses the agent’s ability to identify the correct files that need to be modified within a repository. Let us assume that we would need to modify file.py to solve our problem. If our coding agent implements a change in any other file, it would receive a file retrieval score of 0.
- CST Node-level Retrieval: evaluates the agent’s ability to identify specific code structures that require changes. It uses the Concrete Syntax Tree (CST) representation of the code to measure how accurately the agent can locate the exact functions or classes that need modification.
Figure 2: Illustration of CST node changes.
In Figure 2, we see a change in class node A materialized by a change in its initialization function on the left path starting from the file node. In contrast to the first change, the change in class B is considered a function node change as it doesn’t impact class construction.
Let us assume the change that would solve our problem is the change in the __init__ function. If our coding agent implements the change in my_func, it receives both a class and function node retrieval score of 0.
By combining pass rate assessment with both file-level and CST node-level retrieval metrics, SWE-PolyBench offers a detailed evaluation of AI coding assistants’ capabilities in real-world scenarios. This approach provides deeper insights into how well agents navigate and comprehend complex codebases, going beyond simple task completion to assess their understanding of code structure and organization.
Performance of open-source coding agents
Key Findings
- Language Proficiency: Python is the strongest language for all agents, likely due to its prevalence in training data and existing benchmarks.
- Complexity Challenges: Performance degrades as task complexity increases, particularly when modifications to 3 or more files are required.
- Task Specialization: Different agents show strengths in various task categories (bug fixes, feature requests, refactoring).
- Context Importance: The informativeness of problem statements impacts success rates across all agents (refer to Figure 5 of the appendix paper for details about this analysis).
Many existing open-source agents are designed primarily for Python. Adapting them to work for all four languages of SWE-PolyBench required adjusting test execution commands, modifying parsing mechanisms, and adapting containerization strategies for each language. We adapted and evaluated three open-source agents on SWE-PolyBench. The aforementioned adjustments are reflected by the added “-PB” suffix to the original agent names.
Figure 3: Performance of coding agents across programming languages and task complexities, highlighting strengths and areas for improvement.
Figure 3 provides a visual representation of agent performance across different dimensions:
- Language Proficiency: The left side of the chart shows that all three agents perform best in Python, with significantly lower pass rates in other languages. This highlights the current bias towards Python in many coding agents and their underlying large language models.
- Task Complexity: The right side of the chart illustrates how performance degrades as task complexity increases. Agents show higher pass rates for tasks involving single class or function changes, but struggle with tasks requiring modifications to multiple classes or functions and in instances where both class and function changes are required.
This comprehensive view of agent performance underscores the value of SWE-PolyBench in identifying specific strengths and weaknesses of different coding assistants, paving the way for targeted improvements in future iterations.
In addition to these insights, the evaluation revealed interesting patterns across different task categories as shown in Table 2. The performance data across bug fixes, feature requests, and refactoring tasks reveals varying strengths among AI coding assistants. The performance on bug fixing tasks is relatively consistent. There is more variability between different agents and between multiple runs of a given agent for feature request tasks and refactoring tasks.
Join the SWE-PolyBench community
SWE-PolyBench and its evaluation framework are publicly available. This open approach invites the global developer community to build upon this work and advance the field of AI-assisted software engineering. As coding agents continue to evolve, benchmarks like SWE-PolyBench play a crucial role in ensuring they can meet the diverse needs of real-world software development across multiple programming languages and task types.
Explore SWE-PolyBench today and contribute to the future of AI-powered software engineering!
Resources
Authors
Regulating AI Behavior with a Hypervisor
Post Syndicated from Bruce Schneier original https://www.schneier.com/blog/archives/2025/04/regulating-ai-behavior-with-a-hypervisor.html
Interesting research: “Guillotine: Hypervisors for Isolating Malicious AIs.”
Abstract:As AI models become more embedded in critical sectors like finance, healthcare, and the military, their inscrutable behavior poses ever-greater risks to society. To mitigate this risk, we propose Guillotine, a hypervisor architecture for sandboxing powerful AI models—models that, by accident or malice, can generate existential threats to humanity. Although Guillotine borrows some well-known virtualization techniques, Guillotine must also introduce fundamentally new isolation mechanisms to handle the unique threat model posed by existential-risk AIs. For example, a rogue AI may try to introspect upon hypervisor software or the underlying hardware substrate to enable later subversion of that control plane; thus, a Guillotine hypervisor requires careful co-design of the hypervisor software and the CPUs, RAM, NIC, and storage devices that support the hypervisor software, to thwart side channel leakage and more generally eliminate mechanisms for AI to exploit reflection-based vulnerabilities. Beyond such isolation at the software, network, and microarchitectural layers, a Guillotine hypervisor must also provide physical fail-safes more commonly associated with nuclear power plants, avionic platforms, and other types of mission critical systems. Physical fail-safes, e.g., involving electromechanical disconnection of network cables, or the flooding of a datacenter which holds a rogue AI, provide defense in depth if software, network, and microarchitectural isolation is compromised and a rogue AI must be temporarily shut down or permanently destroyed.
The basic idea is that many of the AI safety policies proposed by the AI community lack robust technical enforcement mechanisms. The worry is that, as models get smarter, they will be able to avoid those safety policies. The paper proposes a set technical enforcement mechanisms that could work against these malicious AIs.
Android Improves Its Security
Post Syndicated from Bruce Schneier original https://www.schneier.com/blog/archives/2025/04/android-improves-its-security.html
Android phones will soon reboot themselves after sitting idle for three days. iPhones have had this feature for a while; it’s nice to see Google add it to their phones.