Daily Tech Digest - January 30, 2025


Quote for the day:

"Uncertainty is not an indication of poor leadership; it underscores the need for leadership." -- Andy Stanley


Doing authentication right

Like encryption, authentication is one of those things that you are tempted to “roll your own” but absolutely should not. The industry has progressed enough that you should definitely “buy and not build” your authentication solution. Plenty of vendors offer easy-to-implement solutions and stay diligently on top of the latest security issues. Authentication also becomes a tradeoff between security and a good user experience. ... Passkeys are a relatively new technology and there is a lot of FUD floating around out there about them. The bottom line is that they are safe, secure, and easy for your users. They should be your primary way of authenticating. Several vendors make implementing passkeys not much harder than inserting a web component in your application. ... Forcing users to use hard-to-remember passwords means they will be more likely to write them down or use a simple password that meets the requirements. Again, it may seem counterintuitive, but XKCD has it right. In addition, the longer the password, the harder it is to crack. Let your users create long, easy-to-remember passwords rather than force them to use shorter, difficult-to-remember passwords. ... Six digits is the outer limit for OTP links, and you should consider shorter ones. Under no circumstances should you require OTPs longer than six digits because they are vastly harder for users to keep in short-term memory.


Augmenting Software Architects with Artificial Intelligence

Technical debt is mistakenly thought of as just a source code problem, but the concept is also applicable to source data (this is referred to as data debt) as well as your validation assets. AI has been used for years to analyze existing systems to identify potential opportunities to improve the quality (to pay down technical debt). SonarQube, CAST SQG and BlackDuck’s Coverity Static Analysis statically analyze existing code. Applitools Visual AI dynamically finds user interface (UI) bugs and Veracode’s DAST to find runtime vulnerabilities in web apps. The advantages of this use case are that it pinpoints aspects of your implementation that potentially should be improved. As described earlier, AI tooling offers to the potential for greater range, thoroughness, and trustworthiness of the work products as compared with that of people. Drawbacks to using AI-tooling to identify technical debt include the accuracy, IP, and privacy risks described above. ... As software architects we regularly work with legacy implementations that they need to leverage and often evolve. This software is often complex, using a myriad of technologies for reasons that have been forgotten over time. Tools such as CAST Imaging visualizes existing code and ChartDB visualizes legacy data schemas to provide a “birds-eye view” of the actual situation that you face.


Keep Your Network Safe From the Double Trouble of a ‘Compound Physical-Cyber Threat'

Your first step should be to evaluate the state of your company’s cyber defenses, including communications and IT infrastructure, and the cybersecurity measures you already have in place—identifying any vulnerabilities and gaps. One vulnerability to watch for is a dependence on multiple security platforms, patches, policies, hardware, and software, where a lack of tight integration can create gaps that hackers can readily exploit. Consider using operational resilience assessment software as part of the exercise, and if you lack the internal know-how or resources to manage the assessment, consider enlisting a third-party operational resilience risk consultant. ... Aging network communications hardware and software, including on-premises systems and equipment, are top targets for hackers during a disaster because they often include a single point of failure that’s readily exploitable. The best counter in many cases is to move the network and other key communications infrastructure (a contact center, for example) to the cloud. Not only do cloud-based networks such as SD-WAN, (software-defined wide area network) have the resilience and flexibility to preserve connectivity during a disaster, they also tend to come with built-in cybersecurity measures.


California’s AG Tells AI Companies Practically Everything They’re Doing Might Be Illegal

“The AGO encourages the responsible use of AI in ways that are safe, ethical, and consistent with human dignity,” the advisory says. “For AI systems to achieve their positive potential without doing harm, they must be developed and used ethically and legally,” it continues, before dovetailing into the many ways in which AI companies could, potentially, be breaking the law. ... There has been quite a lot of, shall we say, hyperbole, when it comes to the AI industry and what it claims it can accomplish versus what it can actually accomplish. Bonta’s office says that, to steer clear of California’s false advertising law, companies should refrain from “claiming that an AI system has a capability that it does not; representing that a system is completely powered by AI when humans are responsible for performing some of its functions; representing that humans are responsible for performing some of a system’s functions when AI is responsible instead; or claiming without basis that a system is accurate, performs tasks better than a human would, has specified characteristics, meets industry or other standards, or is free from bias.” ... Bonta’s memo clearly illustrates what a legal clusterfuck the AI industry represents, though it doesn’t even get around to mentioning U.S. copyright law, which is another legal gray area where AI companies are perpetually running into trouble.


Knowledge graphs: the missing link in enterprise AI

Knowledge graphs are a layer of connective tissue that sits on top of raw data stores, turning information into contextually meaningful knowledge. So in theory, they’d be a great way to help LLMs understand the meaning of corporate data sets, making it easier and more efficient for companies to find relevant data to embed into queries, and making the LLMs themselves faster and more accurate. ... Knowledge graphs reduce hallucinations, he says, but they also help solve the explainability challenge. Knowledge graphs sit on top of traditional databases, providing a layer of connection and deeper understanding, says Anant Adya, EVP at Infosys. “You can do better contextual search,” he says. “And it helps you drive better insights.” Infosys is now running proof of concepts to use knowledge graphs to combine the knowledge the company has gathered over many years with gen AI tools. ... When a knowledge graph is used as part of the RAG infrastructure, explicit connections can be used to quickly zero in on the most relevant information. “It becomes very efficient,” said Duvvuri. And companies are taking advantage of this, he says. “The hard question is how many of those solutions are seen in production, which is quite rare. But that’s true of a lot of gen AI applications.”


U.S. Copyright Office says AI generated content can be copyrighted — if a human contributes to or edits it

The Copyright Office determined that prompts are generally instructions or ideas rather than expressive contributions, which are required for copyright protection. Thus, an image generated with a text-to-image AI service such as Midjourney or OpenAI’s DALL-E 3 (via ChatGPT), on its own could not qualify for copyright protection. However, if the image was used in conjunction with a human-authored or human-edited article (such as this one), then it would seem to qualify. Similarly, for those looking to use AI video generation tools such as Runway, Pika, Luma, Hailuo, Kling, OpenAI Sora, Google Veo 2 or others, simply generating a video clip based on a description would not qualify for copyright. Yet, a human editing together multiple AI generated video clips into a new whole would seem to qualify. The report also clarifies that using AI in the creative process does not disqualify a work from copyright protection. If an AI tool assists an artist, writer or musician in refining their work, the human-created elements remain eligible for copyright. This aligns with historical precedents, where copyright law has adapted to new technologies such as photography, film and digital media. ... While some had called for additional protections for AI-generated content, the report states that existing copyright law is sufficient to handle these issues.


From connectivity to capability: The next phase of private 5G evolution

Faster connectivity is just one positive aspect of private 5G networks; they are the basis of the current digital era. These networks outperform conventional public 5G capabilities, giving businesses incomparable control, security, and flexibility. For instance, private 5G is essential to the seamless connection of billions of devices, ensuring ultra-low latency and excellent reliability in the worldwide IoT industry, which has the potential to reach $650.5 billion by 2026, as per Markets and Markets. Take digital twins, for example—virtual replicas of physical environments such as factories or entire cities. These replicas require real-time data streaming and ultra-reliable bandwidth to function effectively. Private 5G enables this by delivering consistent performance, turning theoretical models into practical tools that improve operational efficiency and decision-making. ... Also, for sectors that rely on efficiency and precision, the private 5G is making big improvements in this area. For instance, in the logistics sector, it connects fleets, warehouses, and ports with fast, low-latency networks, streamlining operations throughout the supply chain. In fleet management, private 5G allows real-time tracking of vehicles, improving route planning and fuel use. 


American CISOs should prepare now for the coming connected-vehicle tech bans

The rule BIS released is complex and intricate and relies on many pre-existing definitions and policies used by the Commerce Department for different commercial and industrial matters. However, in general, the restrictions and compliance obligations under the rule affect the entire US automotive industry, including all-new, on-road vehicles sold in the United States (except commercial vehicles such as heavy trucks, for which rules will be determined later.) All companies in the automotive industry, including importers and manufacturers of CVs, equipment manufacturers, and component suppliers, will be affected. BIS said it may grant limited specific authorizations to allow mid-generation CV manufacturers to participate in the rule’s implementation period, provided that the manufacturers can demonstrate they are moving into compliance with the next generation. ... Connected vehicles and related component suppliers are required to scrutinize the origins of vehicle connectivity systems (VCS) hardware and automated driving systems (ADS) software to ensure compliance. Suppliers must exclude components with links to the PRC or Russia, which has significant implications for sourcing practices and operational processes.


What to know about DeepSeek AI, from cost claims to data privacy

"Users need to be aware that any data shared with the platform could be subject to government access under China's cybersecurity laws, which mandate that companies provide access to data upon request by authorities," Adrianus Warmenhoven, a member of NordVPN's security advisory board, told ZDNET via email. According to some observers, the fact that R1 is open-source means increased transparency, giving users the opportunity to inspect the model's source code for signs of privacy-related activity. Regardless, DeepSeek also released smaller versions of R1, which can be downloaded and run locally to avoid any concerns about data being sent back to the company (as opposed to accessing the chatbot online). ... "DeepSeek's new AI model likely does use less energy to train and run than larger competitors' models," confirms Peter Slattery, a researcher on MIT's FutureTech team who led its Risk Repository project. "However, I doubt this marks the start of a long-term trend in lower energy consumption. AI's power stems from data, algorithms, and compute -- which rely on ever-improving chips. When developers have previously found ways to be more efficient, they have typically reinvested those gains into making even bigger, more powerful models, rather than reducing overall energy usage."


The AI Imperative: How CIOs Can Lead the Charge

For CIOs, AGI will take this to the next level. Imagine systems that don't just fix themselves but also strategize, optimize and innovate. AGI could automate 90% of IT operations, freeing up teams to focus on strategic initiatives. It could revolutionize cybersecurity by anticipating and neutralizing threats before they strike. It could transform data into actionable insights, driving smarter decisions across the organization. The key is to begin incrementally, prove the value and scale strategically. AGI isn't just a tool; it's a game-changer. ... Cybersecurity risks are real and imminent. Picture this: you're using an open-source AI model and suddenly, your system gets hacked. Turns out, a malicious contributor slipped in some rogue code. Sounds like a nightmare, right? Open-source AI is powerful, but has its fair share of risks. Vulnerabilities in the code, supply chain attacks and lack of appropriate vendor support are absolutely real concerns. But this is true for any new technology. With the right safeguards, we can minimize and mitigate these risks. Here's what I recommend: Regularly review and update open-source libraries. CIOs should encourage their teams to use tools like software composition analysis to detect suspicious changes. Train your team to manage and secure open-source AI deployments. 

No comments:

Post a Comment