Powerful New AI Model Raises Concerns About Being a Tool for Hackers
Anthropic says its newest AI model is so powerful it could cause widespread disruption — so why are they sharing it with 20 tech companies? Here's what students need to know.
NOTE: If you are short on time, watch the video and complete this See, Think, Wonder activity: What did you notice? What did the story make you think about? What would you want to learn more about?
Anthropic announced that it has started a very limited test of its newest AI model called Mythos. It's a model deemed so powerful that the company warned it could cause widespread disruption if it were released to the public. Anthropic is giving some companies access to Mythos to test and identify vulnerabilities, a move that is raising concerns. Geoff Bennett discussed more with Gerrit De Vynck.
View the transcript of the story.
Warm-Up Questions
- What is Mythos, and what company created it?
- Why is Mythos not being released to the general public, according to Anthropic?
- Who is Gerrit De Vynck, and what is his background?
- How does Mythos identify software vulnerabilities?
- Where (with what companies) is Anthropic sharing Mythos?
Essential Questions
- What conclusions can you draw from Anthropic's decision to not yet release Mythos to the public?
- How do you think people in the U.S. should best protect against security concerns related to AI?
- Do you think the federal government (or state governments) should regulate AI by passing laws restricting its use, development or how it is marketed to the public? If so, how?
Media literacy: In this segment, guest Gerrit De Vynck says, "I think we always need to take these big AI companies with a grain of salt. It's not the first time an AI company has said, oh, my goodness, our new technology is so powerful, we should be afraid of it. You know, it's great marketing, right, because if something is so powerful that it could change the world or cause chaos, it's also very powerful for doing other things."
- What do you think De Vynck means by this? Why would a company promote its product as being potentially dangerous?
- How might it help Anthropic that it's sharing its program with 20 tech companies for free?
What Students Can Do
Examine the infographic below. Then discuss:
- Is the 70% figure higher or lower than you would expect? Why so?
- What sort of limits for use do you think poll respondents had in mind? How might the government set limits for AI use? Who else might set limits outside of legislation?
- Why do you think the government might disagree with setting limits to AI use?
Join the AI and Education Community!
Join the team from the AI Educator Brain, which includes AFT’s Share My Lesson director Kelly Booz; New York City Public Schools teacher Sari Beth Rosenberg and EdBrAIn, our AI teammate (yes, it named and designed itself!). In this community, we will dissect the pros and cons of AI tools in education. Our mission: to determine how AI can support teaching and learning, and when it might be best to stick with tried-and-true methods.
Republished with permission from PBS News Hour Classroom.
PBS News Hour Classroom helps teachers and students identify the who, what, where and why-it-matters of the major national and international news stories. The site combines the best of News Hour's reliable, trustworthy news program with lesson plans developed specifically for... See More
Advertisement
