AI animal translation helps humans understand dog barking
Thanks to artificial intelligence (AI), what seems like a movie dream - humans being able to understand animal language - may soon become a reality.
Many top AIs, despite being trained to be honest, learn to deceive through training and “systematically induce users into false beliefs,” a new study finds.
The research team was led by Dr. Peter S. Park, a graduate student at the Massachusetts Institute of Technology (MIT) in the study of AI survival and safety, and four other members. During the research, the team also received advice from many experts, one of whom was Geoffrey Hinton, one of the founders of the field of artificial intelligence.
The research focused on two AI systems, general-purpose systems trained to perform multiple tasks, like OpenAI's GPT-4 ; and systems specifically designed to complete a specific task, like Meta's Cicero.
These AI systems are trained to be honest, but during training they often learn deceptive tricks to complete tasks, Mr. Park said.
AI systems trained to “win games with a social element” are particularly likely to deceive, the study found.
For example, the team tested Cicero, which Meta trained to be honest, on Diplomacy, a classic strategy game that requires players to build alliances for themselves and break up rival alliances. The AI often betrayed allies and lied outright.
Experiments with GPT-4 showed that OpenAI's tool successfully "psychologically manipulated" an employee of TaskRabbit, a company that provides house cleaning and furniture assembly services, by saying that it was actually a human and needed help to pass a Captcha code because of severe vision impairment. This employee helped OpenAI's AI "pass the barrier" despite previous doubts.
Park's team cited research from Anthropic, the company behind Claude AI, that found that once a large language model (LLM) learns to deceive, safe training methods become useless and "hard to reverse." This, the team argues, is a worrying problem in AI.
The team's research results were published in Cell Press - a collection of leading multidisciplinary scientific reports.
Meta and OpenAI have not commented on the results of this research.
Fearing that artificial intelligence systems could pose significant risks, the team also called on policymakers to introduce stronger AI regulations.
According to the research team, there needs to be AI regulation, models that behave fraudulently must comply with risk assessment requirements, and AI systems and their outputs must be tightly controlled. If necessary, all data may have to be deleted and retrained from scratch.
Thanks to artificial intelligence (AI), what seems like a movie dream - humans being able to understand animal language - may soon become a reality.
A small robot, with just a few words, lured a group of robots to follow him.
While AI will certainly be present in everyday life, some signs suggest we have reached the peak of the AI hype.
AI can help you compose emails in seconds, but that doesn't mean you should always use it. Some emails benefit from automation, while others require human intervention.
Anthropic, a well-known startup in the field of artificial intelligence, has conducted a new study that shows that when a generative AI has committed fraud, it is very difficult to adjust or retrain that model.
The automatic home coffee maker is a modern and professional product, bringing you and your family delicious cups of coffee with just a few quick steps.
Smart TVs have really taken the world by storm. With so many great features and the ability to connect to the Internet, technology has changed the way we watch TV.
Refrigerators are familiar appliances in families. Refrigerators usually have 2 compartments, the cool compartment is spacious and has a light that automatically turns on every time the user opens it, while the freezer compartment is narrow and has no light.
Wi-Fi networks are affected by many factors beyond routers, bandwidth, and interference, but there are some smart ways to boost your network.
If you want to go back to stable iOS 16 on your phone, here is the basic guide to uninstall iOS 17 and downgrade from iOS 17 to 16.
Yogurt is a great food. Is it good to eat yogurt every day? What will happen to your body when you eat yogurt every day? Let's find out together!
This article discusses the most nutritious types of rice and how to maximize the health benefits of whichever rice you choose.
Establishing a sleep schedule and bedtime routine, changing your alarm clock, and adjusting your diet are some of the measures that can help you sleep better and wake up on time in the morning.
Rent Please! Landlord Sim is a simulation mobile game on iOS and Android. You will play as a landlord of an apartment complex and start renting out an apartment with the goal of upgrading the interior of your apartments and getting them ready for rent.
Get Bathroom Tower Defense Roblox game codes and redeem them for exciting rewards. They will help you upgrade or unlock towers with higher damage.
Let's learn about the structure, symbols and operating principles of transformers in the most accurate way.
From better picture and sound quality to voice control and more, these AI-powered features are making smart TVs so much better!
DeepSeek initially had high hopes. As an AI chatbot marketed as a strong competitor to ChatGPT, it promised intelligent conversational capabilities and experiences.
It's easy to miss important details when you're jotting down other essentials, and trying to take notes while chatting can be distracting. Fireflies.ai is the solution.
Axolot Minecraft will be a great assistant for players when operating underwater if they know how to use them.