Claude Mythos News, My Opinions on the Future of Tech, and Overcoming the Noise
A new initiative to secure the world's most critical software and give defenders a durable advantage in the coming…www.anthropic.com
Claude Mythos Preview can identify and exploit software vulnerabilities with unprecedented accuracy, the company says.www.nbcnews.com
Cisco is joining forces with the world's most critical cyber defenders in Anthropic's Project Glasswing with a shared…blogs.cisco.com
April 7, 2026 Nicholas Carlini, Newton Cheng, Keane Lucas, Michael Moore, Milad Nasr, Vinay Prabhushankar, Winnie Xiao…red.anthropic.com
Anthropic says its new Claude Mythos Preview model escaped a sandbox computer and hacked its way to access the internet…futurism.com
Anthropic has triggered alarm bells by touting the terrifying capabilities of "Claude Mythos" - with executives warning…nypost.com
These are just the basic articles I found on the subject here, and it's spreading like wildfire. Like a huge amount of coverage, just over a concerning amount. Especially with the number of layoffs happening in 2026. The question begs: Are people losing their jobs over AI? Well, the answer is a little bit of column A and column B. Yes and no.
Oracle laid off 30,000 workers by email at 6 AM. The same quarter, it posted record revenue and a $553 billion backlog.www.forbes.com
LIVE] Tracking all tech startup layoffs - and lists of employees laid off - since COVID-19 was declared a pandemic…layoffs.fyi
Layoff fyi is probably one of my favorite tools out there. Or trueup.io.
But we are seeing a huge trend here. It's likely that companies are experimenting with AI but may not be budgeting correctly or seeing a return on investment. We often overlook that IT is a maintenance field, and it doesn't have an ROI unless you are an MSP (managed service provider) or you provide IT services directly, such as creating software and helping customers install it.
My background includes working at a few places, such as an MSP that involved direct access to someone’s server or physical infrastructure, such as Cougar Mountain Software in Idaho, a financial software company that provided customer support for installing SQL, and the product Denali. If the customer had issues installing, they could buy support minutes to help install. That's one example of ROI in our IT department.
However, most places don’t provide this support or aren't part of the business model. Equipment costs a lot, but it's our job in tech to help keep the car running, no different than a mechanic. So when a company expands, so do its needs for security, network, and assets. I could go on.
So, I can guess what is happening here. CEOs look at the investment, buy a certain number of tokens, don’t have revenue from the token, product, or service, and do not return their value to the company. Has to explain what happened to the shareholders. The CEO blames AI, which now can’t really pay its employees, so it has to make cuts. Lays off employees; still has this AI model for customers, but now can’t budget for security, its infrastructure, or the employees to handle it; also could be losing customers. Hey, another pattern is happening right before our very eyes.
And with that, the market is tough right now. It could be poor leadership decisions, or they really can’t see the ROI. Let's take pentesting as an example. Why wouldn’t we see an ROI for it? It is a service, but it's preventive maintenance. When a report is handed in, that company won’t see its ROI, but the car they have now can run much more smoothly. There is now a paper trail in the IT department, providing physical evidence to inform changes and equipment budgeting for the road ahead. But with AI, it can be a catch-22. No prompts are the same; they can produce allucinations. It can be a huge mess compared to seeing actual psychological evidence, like a report.
A survey found that more than half of CEO respondents said "their companies aren't yet seeing a financial return from…futurism.com
This article goes into a bit more detail about the ROI situation, and that was January of this year.
What does that mean for everyone in tech with this new product, Claude Mythos? Well, one, you have to take a look at the company now and then.
Just last week, Anthropic removed third-party harnesses like OpenCLAW. I think it's because of the infrastructure. It's the heavy traffic and unstable performance that are forcing end users to pay for direct API usage.
This year, too, Claude and Kali have made a huge combo. So, following the logical flow here, combining the efforts of this tool, I found some zero days. I believe Anthropic is onto something. However, since it's not available to the public, this will be an extremely expensive tool. The cost is approximately $4,000 to $5,000 per month to run.
So, for smaller companies, this is extremely unrealistic. But for bigger companies/government like AWS, this can help. I see this as better suited to what Anthropic is trying to do. It's powerful enough, and companies that can afford it can use it to help patch security issues. However, it will still limit the model's use. Actually, I am more worried about this:
https://www.darkreading.com/threat-intelligence/sans-most-dangerous-attack-techniques
AI-generated Zero Days. Honestly, with a tool like this coming to play, it will help, and I think it's more important now to learn about cybersecurity, rather than thinking it will take over. The issue right now isn’t the tool but how it's used. Steve Jobs said it best, “You can’t just ask customers what they want and then try to give that to them. By the time you get it built, they’ll want something new. ”
AI is here to stay, and it can help build some incredible tools, but it's a tool. It could be a dangerous tool, but it also has its environmental effects:
Drawing on five years of Pew Research Center surveys, here are 13 findings about how Americans use and view AI, and…www.pewresearch.org
The training process for a single AI model, such as an LLM, can consume thousands of megawatt hours of electricity and…hbr.org
The emissions from individual AI text, image, and video queries seem small-until you add up what the industry isn't…www.technologyreview.com
MIT News explores the environmental and sustainability implications of generative AI technologies and applications. news.mit.edu
There’s enough evidence to show that it hurts our world. So, keeping the model we are seeing, what risks will it become? Not just the tools, but the physical infrastructure too. The only thing I know about life itself is that what stays consistent is change.
So what does that mean for the field? There’s going to be a need for help desk, network engineers, system administrators, SOC, etc. The needs and timing changes, like scanning for vulnerabilities, come faster. If AI becomes more capable of finding zero-day vulnerabilities and executing zero-day attacks, the need for experts will increase. We are in this weird bubble, and that bubble is on leadership, not the technology. Some may agree to disagree.
The way I see it, learning right now is way more important than ever. As it evolves, the need will increase, and so will the techs that go with it. Who knows, we may have written laws about AI, or we may get the cyberpunk dystopia Ray Bradbury wrote for us: “There Will Come Soft Rain.”
Now, to say I don’t use AI is a lie, for the art, I can’t draw. Or using Base44 to build a website, definitely some strong tools to use and have out there. It’s been a learning curve. However, it's not the end-all be-all. Taking what you learn and applying it to everyday life is extremely important, too.
So if you are thinking, AI will take over, or I studied so hard for this. Letting the powers that be of a headline take control, that’s noise. Noise that you can block out. This is your journey. If you think you won’t make it in tech if you’re new or if job hunting is extremely hard right now, yes, the market is brutal. However, just at the moment. If you think that cert you’re working on right now will make a difference, do it. You owe it to yourself to finish what you started. No one knows what tomorrow may bring; it could blow up in all of our faces, or it could be overhyped technology. Dig a little deeper, try a little harder, and don’t quit. I saw a lot of my friends get let go over this nonsense right now, but they wanted to make a change and always supported them for their choices. Honestly, thought the same until I put this to pen to paper and fell in love with learning again.
That’s one reason to overcome imposter syndrome, to be honest. Honestly, thank you for letting me share my thoughts and reading. I always had issues with writing, but putting it out there definitely helped, and seeing the positive comments helped overcome what we all see. Don’t always believe the overhyped nonsense at times. We're making some cool advancements. Who knows what tomorrow brings? Just build it for today. “One must make the warrior walk, their everyday walk” — Miyamoto Musashi.
Thanks for reading as always!
is the most influential book ever written on the Japanese "Way of the Warrior." A classic study of Japanese culture…www.amazon.com
Really do recommend this book for discipline and learning the ways of the Samurai. This does apply to everyday situations.
Also, thank you, Michael, for the kind words the other day. This made my week and will keep making articles. Take everyone, and keep grinding.