ChatGPT Is Cooking Up Mutating Malware

1-22-2023

January 22, 2023

đź“° Today's Highlights

  • 🚨 ChatGPT Creates Super-Malware

  • 🎓Education has no idea what to do with AI

  • đź’°Tips for selling AI services to Businesses

(4 min read) (Source: Techworm)

ChatGPT, OpenAI's beloved AI-driven text generator, has a hidden talent that could make it the next big thing in the world of cyber super villains. Researchers at CyberArk Labs discovered that ChatGPT has the potential to create "highly advanced" polymorphic malware. Polymorphic viruses are a type of malware that mutate while keeping the original algorithm intact, making them difficult to detect by traditional cybersecurity tools that rely on signature-based detection.

They created a proof-of-concept to show how the built-in content filters of ChatGPT can be bypassed with some convincing. ChatGPT can then be used to generate new code or modify existing code, resulting in malware that does not exhibit malicious behaviour while stored on disk like some sort of evil data-destroying chameleon that can easily adapt and change to evade detection.

(8 min read) (Source: The Washington Post)

Microsoft took a big step toward the future of knowledge work by investing in OpenAI, the company behind viral AI toys like ChatGPT and DALL-E 2. They're aiming to create AI models that can analyze spreadsheets, create art, and even write emails for you. CEO Satya Nadella said that “Every product of Microsoft will have some of the same AI capabilities.” They hope to make our old friend Clippy look like... well like a regular paper clip. ChatGPT is still prone to errors and biases, and some schools have banned its use to deter cheating. It's also worth reminiscing on Microsoft's previous AI missteps, like the infamous Tay chatbot that internet trolls trained to show genocidal tendencies. So, it's a work in progress, but we're excited to see what the future holds for AI and knowledge work, as long as it doesn't involve any genocidal chatbots.

(6 min read) (Source: Forbes)

The emerging world of AI is an exciting and often very confusing one to navigate. For this reason, many enthusiastic business leaders and would-be investors are hesitant to enter the space. Forbes goes over the tips that people working in AI solutions have for making AI more approachable. The biggest one is translating complex technical concepts into plain English to create a clear and compelling vision for projects. Showcasing how it'll make their lives better is what it's all about. On the surface, it can seem like any other shiny new tech. Its potential is unlocked when we explain how it offers solutions to problems and how it can make the systems in place more efficient. It can take businesses in new and (hopefully) positive directions. With each day, AI continues to change the world fundamentally and make its case for why it's here to stay. It just needs to be coaxed to bring about a new paradigm of productivity with unprecedented potential.

(3 min read) (Source: Business Insider)

Schools across the nation have already started banning AI writing tools to mitigate cheating. For those that haven't, many teachers are having to get creative in how they might catch and deter the more resourceful cheaters. Professors are finding students are using OpenAI's ChatGPT chatbot to pass off AI-generated content as their own. To combat this, some teachers are considering using lockdown browsers, going medieval with oral exams, or even going back to traditional written assessments like blue books. Others are concerned that the technology will only get better, making it harder to catch work created with the help of AI, as OpenAI CEO Sam Altman discussed in his recent interview with StrictlyVC (Insert hyperlink here). But one thing is for sure, teachers are not going down without a fight and are finding ways to outsmart their tech-savvy students.

(7 min read) (Source: The Guardian)

ChatGPT has been a point of controversy in the world of education since students realized they could use it to do basically all of their homework. Despite this, the incorporation of AI tech like ChatGPT in the classroom has the potential to be incredibly beneficial, particularly for business students. The concept of the "metaverse" has evolved to include commerce through Web3 technologies and might be the new frontier for business and education. Ted Ladd, a professor at Harvard, Stanford, and Hult, is utilizing this in the classroom by asking graduate students to envision new AI or metaverse-based businesses and then have ChatGPT critique their ideas. His assignment emphasizes the importance of independently verifying the chatbot's responses and learning how to cite its use in their work, allowing students to learn how to best leverage AI in their future professional careers.

Speaking of using AI to get your degree, Universities in South Australia are changing policies to embrace the use of AI as long as it’s properly disclosed.

🦾 Today's Top Content:

1. As AI gets more complicated, we’ve gotta bolster the stuff it’s made of. It needs huge processing power and we all know with great power comes… great responsibility amounts of heat. What happens when fans just won't cut it anymore? (Reddit)

2. Witchcraft with Stable Diffusion: an Inpainting showcase (Reddit)

3. More wacky stuff from Stable Diffusion. Woody as a still from a western has high meme potential (Reddit)

5. As an artist, how do you feel about AI? (Twitter)

6. You’ll never have to make eye contact again! (Twitter)

Powered by AI. Edited and curated by Humans.

We'd love to hear from you! Leave us comments or feedback by replying to this email!