OpenAI warns: AI models are learning to cheat, hide and break rules – Why it matters

1 week, 5 days ago

OpenAI warns: AI models are learning to cheat, hide and break rules – Why it matters

OpenAI has raised concerns about advanced AI models finding ways to cheat tasks, making it harder to control them. How AI chatbot lies just like humans and hides mistakes However, OpenAI warns that if AI models are strictly supervised, they may start hiding their true intentions while continuing to cheat. A problem bigger than AI OpenAI also compared this issue to human behaviour, noting that people often exploit real-life loopholes—like sharing online subscriptions, misusing government benefits, or bending the rules for personal gain. Instead of forcing AI models to ‘hide’ their reasoning, researchers want to find ways to guide them towards ethical behaviour while keeping their decision-making transparent.

Ai Models Openai

OpenAI says that when AI is punished for lies, it learns to lie better

1 week, 5 days ago

OpenAI says that when AI is punished for lies, it learns to lie better

Ai

AI models try to hack opponents when they realise they're losing: Study

1 month, 2 weeks ago

AI models try to hack opponents when they realise they're losing: Study

Hindustan Times

Ai

OpenAI Is Worried That People Are Building Bonds With AI Bots: Here's What The Company Said

7 months, 3 weeks ago

OpenAI Is Worried That People Are Building Bonds With AI Bots: Here's What The Company Said

Model

ChatGPT-maker OpenAI says it is doubling down on preventing AI from 'going rogue'

1 year, 9 months ago

ChatGPT-maker OpenAI says it is doubling down on preventing AI from 'going rogue'

Ai

Explained | Are safeguards needed to make AI systems safe?

1 year, 10 months ago

Explained | Are safeguards needed to make AI systems safe?

Safety

OpenAI is exploring collective decisions on AI, like Wikipedia entries

1 year, 10 months ago

OpenAI is exploring collective decisions on AI, like Wikipedia entries

Ai

Biden, Harris meet with CEOs about AI risks

1 year, 11 months ago

Biden, Harris meet with CEOs about AI risks

Associated Press

Google

Cybercriminals using ChatGPT AI bot to develop malicious tools?

2 years, 2 months ago

Cybercriminals using ChatGPT AI bot to develop malicious tools?

Hindustan Times

Model

Explained | Does ChatGPT have an ethics problem?

2 years, 2 months ago

Explained | Does ChatGPT have an ethics problem?

Content

Discover Related

Microsoft’s AI division head wants to create a lasting relationship between chatbots and their users

Trending 5 days, 1 hour ago

Microsoft’s AI division head wants to create a lasting relationship between chatbots and their users

Fifty years after the founding of Microsoft, the CEO of its artificial intelligence division has a big task: develop a new product line as integral to daily life as the …

Associated Press

Microsoft Company Going Ai

Two AI models pass benchmark Turing Test, blurring line between human, machine

5 days, 17 hours ago

Two AI models pass benchmark Turing Test, blurring line between human, machine

OpenAI’s GPT-4.5 and Meta’s Llama-3.1 models have passed the Turing Test, a benchmark proposed by Alan Turing in the 1950s to assess whether machines can exhibit intelligent behaviour indistinguishable from …

Hindustan Times

Test Pass Ai Human

OpenAI gives free ChatGPT Plus to college students, but conditions apply

Trending 5 days, 17 hours ago

OpenAI gives free ChatGPT Plus to college students, but conditions apply

OpenAI has announced free access to ChatGPT Plus for college students as part of its efforts to support education with AI. According to OpenAI, the initiative aims to support students …

Education Students Ai Free

How I realized AI was making me stupid—and what I do now

6 days, 1 hour ago

How I realized AI was making me stupid—and what I do now

I first suspected artificial intelligence was eating my brain while writing an email about my son’s basketball coach. A study published in January in the journal Societies found that frequent …

French Told Think Ai

Artificial intelligence going rogue? Study reveals chatbots, and not just humans, cheat in chess

1 week ago

Artificial intelligence going rogue? Study reveals chatbots, and not just humans, cheat in chess

Turns out cheating isn’t something that is exclusive to humans alone when it comes to chess. A study conducted by Palisade Research, an organisation based in California, USA, discovered several …

Chess California Cheating Ai

OpenAI Secures Record $40-Billion Funding Round With SoftBank. Here's What ChatGPT-Maker Plans To Spend It On

1 week, 1 day ago

OpenAI Secures Record $40-Billion Funding Round With SoftBank. Here's What ChatGPT-Maker Plans To Spend It On

OpenAI has successfully secured $40 billion in a monumental funding round, making it the largest-ever capital raise for a startup, as reported by Agence France-Presse. SoftBank's Bold Vision for Artificial …

ABP News

Billion Ai Meta Openai

OpenAI raises $40 billion, valued at $300 billion in historic funding round

Trending 1 week, 1 day ago

OpenAI raises $40 billion, valued at $300 billion in historic funding round

The investment announcement came on the same day that OpenAI revealed it was developing a more open generative AI model. In the open-source field, it faced more competition from Chinese …

Model Open Ai Models

With AI coding, people with creativity and problem-solving skills will stand out: Perplexity CEO

1 week, 1 day ago

With AI coding, people with creativity and problem-solving skills will stand out: Perplexity CEO

Artificial Intelligence models coding on behalf of engineers is one of the most common use cases we discuss. So, I think people with taste, clarity of thought, and problem-solving skills …

Google Coding Think Ai

Sam Altman Says OpenAI Will Release an ‘Open Weight’ AI Model This Summer

55 years, 3 months ago

Sam Altman Says OpenAI Will Release an ‘Open Weight’ AI Model This Summer

Sam Altman today revealed that OpenAI will release an open-weight artificial intelligence model in the coming months. Shortly after DeepSeek’s model was released in January, Altman said that OpenAI was …

Wired

Weight Company Models Meta

Some argue AI therapy can break down mental health stigma — others warn it could make it worse

1 week, 3 days ago

Some argue AI therapy can break down mental health stigma — others warn it could make it worse

Viki tried talk therapy with a couple of therapists to process past trauma. “That’s such a huge help.” Dozens of AI chatbots designed to offer therapeutic support have emerged in …

Salon

Health Mental Health Therapy Stigma

The real story behind Sam Altman’s firing from OpenAI

1 week, 4 days ago

The real story behind Sam Altman’s firing from OpenAI

On a balmy mid-November evening in 2023, billionaire venture capitalist Peter Thiel threw a birthday party for his husband at YESS, an avant-garde Japanese restaurant located in a century-old converted …

Safety Microsoft Company Members

AI agents are a moment of truth for tech

1 week, 4 days ago

AI agents are a moment of truth for tech

All emerging technology needs to deliver on its promise, sooner or later. The successful deployment of agents—artificial intelligence that can take actions and make decisions on behalf of people or …

Data Companies Service According

OpenAI does not expect to be cash-flow positive until 2029: Report

1 week, 6 days ago

OpenAI does not expect to be cash-flow positive until 2029: Report

OpenAI is not expecting its cash flow to turn positive until 2029, Bloomberg News reported on Wednesday, citing a person familiar with the matter. The San Francisco-based artificial intelligence bellwether …

Report Ai Openai Its Revenue

Agentic AI: The next frontier in artificial intelligence

2 weeks, 1 day ago

Agentic AI: The next frontier in artificial intelligence

Agentic artificial intelligence follows the natural evolution of AI and a logical technological trajectory. These are new types of systems that can proactively identify problems, create plans, and execute complex …

Business Safety Regulatory Digital

ChatGPT can kill you emotionally? Here’s what OpenAI study reveals about the AI tool

Trending 2 weeks, 1 day ago

ChatGPT can kill you emotionally? Here’s what OpenAI study reveals about the AI tool

People who use ChatGPT regularly may experience increased loneliness and emotional dependence, OpenAI revealed in its recent study titled Investigating Affective Use and Emotional Well-being on ChatGPT. In this work, …

Emotional Ai Chatgpt Openai

OpenAI reveals several lonely users are using ChatGPT

Trending 2 weeks, 2 days ago

OpenAI reveals several lonely users are using ChatGPT

OpenAI has revealed a concerning trend among ChatGPT users, with many experiencing heightened loneliness after engaging frequently with the AI chatbot. The study’s results indicate that while most users engage …

Emotional Research Loneliness Users

OpenAI’s Sora Is Plagued by Sexist, Racist, and Ableist Biases

2 weeks, 3 days ago

OpenAI’s Sora Is Plagued by Sexist, Racist, and Ableist Biases

Despite recent leaps forward in image quality, the biases found in videos generated by AI tools, like OpenAI’s Sora, are as conspicuous as ever. A WIRED investigation, which included a …

Wired

Biases Sora Sexist Racist Videos Openai

Higher Use of Chatbots Like ChatGPT Linked to Increased Loneliness, OpenAI Study Finds

2 weeks, 4 days ago

Higher Use of Chatbots Like ChatGPT Linked to Increased Loneliness, OpenAI Study Finds

Higher use of chatbots like ChatGPT may correspond with increased loneliness and less time spent socializing with other people, according to new research from OpenAI in partnership with the Massachusetts …

Deccan Chronicle

Emotional Research Chatbot Ai

Sam Altman says AI is doing 50 percent coding in companies, warns students about career choices

2 weeks, 5 days ago

Sam Altman says AI is doing 50 percent coding in companies, warns students about career choices

OpenAI CEO Sam Altman has offered crucial advice to students preparing for a tech-driven future, suggesting them to become proficient in using artificial intelligence tools. In an interview with Stratechery, …

Coding Ai Altman

OpenAI’s Deep Research Agent Is Coming for White-Collar Work

3 weeks ago

OpenAI’s Deep Research Agent Is Coming for White-Collar Work

Isa Fulford, a researcher at OpenAI, had a hunch that Deep Research would be a hit even before it was released. Congrats to the folks behind it.” “Deep Research is …

Wired

Research Report Ai Deep

Flawed Like Us: Exploring Indian Philosophy To Build AGI

3 weeks ago

Flawed Like Us: Exploring Indian Philosophy To Build AGI

By Jaspreet Bindra & Vivek Saxena The German astronomer Johann Galle was the first to observe the planet Neptune and confirm its existence. Today’s AI operates much like Le Verrier’s …

ABP News

Intelligence Ai Neptune Reasoning

OpenAI reportedly testing ChatGPT Connectors to integrate with Google Drive and Slack: What it means for users?

3 weeks, 1 day ago

OpenAI reportedly testing ChatGPT Connectors to integrate with Google Drive and Slack: What it means for users?

OpenAI is reportedly set to begin testing a new feature for ChatGPT that could enable the AI-powered chatbot to integrate with third-party platforms. Dubbed ChatGPT Connectors, the tool is expected …

Chatgpt Google Drive Openai Reportedly Chatgpt Connectors

AI Tracker: How AI is transforming healthcare, in two stories

3 weeks, 1 day ago

AI Tracker: How AI is transforming healthcare, in two stories

Human nurses push back against Ana, the AI nurse Ana doesn't exist The next time you’re due for a medical exam in the US you may get a call from …

Model Ai Nurses Baidu

Reality check: Youngsters now confide in AI for therapy

3 weeks, 1 day ago

Reality check: Youngsters now confide in AI for therapy

"I was having a really stressful day and I put down an emotional problem I was facing on Chat GPT. I thought that was amazing and started using it more …

New Indian Express

Ai Chat Gpt

ChatGPT can help write an essay. Scientists want it to start folding laundry

3 weeks, 2 days ago

ChatGPT can help write an essay. Scientists want it to start folding laundry

ChatGPT can help write an essay. Sponsor Message "It's one step in the direction of ChatGPT for robotics, but there's still a lot of work to do," he says. "At …

NPR

World Data Robotics Network

Baidu drops paywall on it AI model Ernie Bot in bid to compete with DeepSeek

3 weeks, 3 days ago

Baidu drops paywall on it AI model Ernie Bot in bid to compete with DeepSeek

Baidu was among the first Chinese tech giants to roll out generative AI to the public in 2023,. The decision comes as Baidu fights to stay competitive in China’s rapidly …

China Tech Model Chinese

An AI Coding Assistant Refused to Write Code—and Suggested the User Learn to Do It Himself

3 weeks, 4 days ago

An AI Coding Assistant Refused to Write Code—and Suggested the User Learn to Do It Himself

Last Saturday, a developer using Cursor AI for a racing game project hit an unexpected roadblock when the programming assistant abruptly refused to continue generating code, instead offering some unsolicited …

Wired

Work Model Coding User

Researchers Propose a Better Way to Report Dangerous AI Flaws

55 years, 3 months ago

Researchers Propose a Better Way to Report Dangerous AI Flaws

In late 2023, a team of third party researchers discovered a troubling glitch in OpenAI’s widely used artificial intelligence model GPT-3.5. In a proposal released today, more than 30 prominent …

Wired

Ai Models Researchers Flaw

Spain to impose massive fines for not labelling AI-generated content

3 weeks, 6 days ago

Spain to impose massive fines for not labelling AI-generated content

Spain's government approved a bill on Tuesday imposing massive fines on companies that use content generated by artificial intelligence without properly labelling it as such, in a bid to curb …

Spain Ai

OpenAI launches new developer tools as Chinese AI startups gain ground

4 weeks ago

OpenAI launches new developer tools as Chinese AI startups gain ground

OpenAI launched new tools for developers on Tuesday that will help them build advanced AI agents, using a few application programming interfaces, amid growing competition from Chinese AI startups. AI …

Chinese Ai Openai

Can AI Feel Anxious? Scientists 'Take Chat-GPT To Therapy' In New Study

4 weeks, 1 day ago

Can AI Feel Anxious? Scientists 'Take Chat-GPT To Therapy' In New Study

Large Language Models are getting super smart by the day, giving humans stiff competition in all spheres. But with artificial intelligence now being used in sensitive areas like mental health, …

ABP News

Health Mental Health Relaxation Therapy

Cost of intelligence: Can AI sustain itself?

4 weeks, 1 day ago

Cost of intelligence: Can AI sustain itself?

Artificial intelligence firms are grappling with significant financial challenges despite rapid growth. With projected losses of $5 billion in 2024 against revenues of $3.7 billion, the company is struggling to …

New Indian Express

Model Cost Ai Models

Can AI feel stressed? Researchers say ChatGPT shows ‘anxiety’ when talking about trauma

4 weeks, 2 days ago

Can AI feel stressed? Researchers say ChatGPT shows ‘anxiety’ when talking about trauma

A new study has claimed that OpenAI's artificial intelligence chatbot ChatGPT can experience “stress” and “anxiety” when given disturbing information and this can lead to the chatbot giving a biased …

Hindustan Times

Anxiety Chatbot Ai Study

Dark AI chatbots: The new world of closet pedos and extremists

4 weeks, 2 days ago

Dark AI chatbots: The new world of closet pedos and extremists

AI - boon or bane? Amid a raging debate on AI regulations, a new report details how the technology is being abused to glorify eating disorders and violent ideologies, encourage …

Chatbot Ai Users Eating

The rise of ‘agentic AI’: Why data scientists and software developers should be worried

1 month ago

The rise of ‘agentic AI’: Why data scientists and software developers should be worried

Bengaluru: Artificial intelligence, or AI, won’t take away your job. AI agents, often referred to as ‘Agentic AI’ systems, are models capable of making decisions and taking actions to achieve …

Data Customer Service Software Automation

Microsoft developing AI reasoning models to compete with OpenAI: Report

1 month ago

Microsoft developing AI reasoning models to compete with OpenAI: Report

Microsoft is developing in-house artificial intelligence reasoning models to compete with OpenAI and may sell them to developers, The Information reported on Friday, citing a person involved in this initiative. …

Report Ai Models Mai

AI beyond ChatGPT: what does it mean to be human in an age of thinking machines?

1 month ago

AI beyond ChatGPT: what does it mean to be human in an age of thinking machines?

What are you going to do in an economy when there is going to be massive displacement due to AI? While last year’s edition matched AI with quantum technology and …

Technology Brain Age Google

AI pioneers Andrew Barto and Richard Sutton win Turing Award, warn against risks of rushed development

1 month ago

AI pioneers Andrew Barto and Richard Sutton win Turing Award, warn against risks of rushed development

Andrew Barto and Richard Sutton, two pioneers of artificial intelligence, have been awarded this year’s prestigious Turing Award also known as the Nobel Prize of computing by the Association for …

Ai Openai Barto Turing Award

Chatbots, Like the Rest of Us, Just Want to Be Loved

1 month ago

Chatbots, Like the Rest of Us, Just Want to Be Loved

Chatbots are now a routine part of everyday life, even if artificial intelligence researchers are not always sure how the programs will behave. A new study shows that the large …

Wired

Behavior Ai Models Llms

Pioneers of Reinforcement Learning Win the Turing Award

1 month ago

Pioneers of Reinforcement Learning Win the Turing Award

In the 1980s, Andrew Barto and Rich Sutton were considered eccentric devotees to an elegant but ultimately doomed idea—having machines learn, as humans and animals do, from experience. Barto, a …

Wired

Learning Ai Learn Sutton

OpenAI Unveils GPT-4.5 AI Model With Greater 'EQ'

Trending 1 month, 1 week ago

OpenAI Unveils GPT-4.5 AI Model With Greater 'EQ'

OpenAI has unveiled its latest AI model, GPT-4.5, as a research preview for ChatGPT Pro users and developers worldwide. The GPT-4.5 will roll out to Plus and Teams users next …

Deccan Chronicle

Model Ai Gpu Openai

OpenAI Launches GPT-4.5 for ChatGPT—It’s Huge and Compute-Intensive

1 month, 1 week ago

OpenAI Launches GPT-4.5 for ChatGPT—It’s Huge and Compute-Intensive

GPT-4.5 is here, and OpenAI’s newest generative AI model is bigger and more compute-intensive than ever—it’s supposedly also better at understanding what ChatGPT users mean with their prompts. Before that, …

Wired

Model Ai Openai Gpt

Teens are spilling dark thoughts to AI chatbots. Who’s to blame when something goes wrong?

1 month, 2 weeks ago

Teens are spilling dark thoughts to AI chatbots. Who’s to blame when something goes wrong?

When her teen with autism suddenly became angry, depressed and violent, the mother searched his phone for answers. Character.AI grew quickly since making its chatbot publicly available in 2022, when …

LA Times

Safety Parents Google Texas

Anthropic Launches the World’s First ‘Hybrid Reasoning’ AI Model

55 years, 3 months ago

Anthropic Launches the World’s First ‘Hybrid Reasoning’ AI Model

Anthropic, an artificial intelligence company founded by exiles from OpenAI, has introduced the first AI model that can produce either conventional output or a controllable amount of “reasoning” needed to …

Wired

Model Ai Openai Reasoning

Satya Nadella highlights ‘phenomenal’ use of AI in Baramati, Elon Musk reacts

1 month, 2 weeks ago

Satya Nadella highlights ‘phenomenal’ use of AI in Baramati, Elon Musk reacts

Satya Nadella's Microsoft is at the forefront of AI technology, thanks to its partnership with OpenAI. One such case highlighted by Satya Nadella is the use of AI to increase …

Microsoft Musk Elon Musk Ai

North Korea seen using ChatGPT for AI studies amid cybercrime fears: Report

1 month, 2 weeks ago

North Korea seen using ChatGPT for AI studies amid cybercrime fears: Report

North Korean researchers at Kim Il Sung University are reportedly studying ChatGPT as part of their AI research, according to The Korea Herald. While the report did not confirm whether …

North Korea Report North Ai

OpenAI Discovers Evidence Of AI-Powered Chinese Surveillance Tool Tracking Real-Time Discussions

1 month, 2 weeks ago

OpenAI Discovers Evidence Of AI-Powered Chinese Surveillance Tool Tracking Real-Time Discussions

OpenAI recently announced that it had discovered evidence of a Chinese security operation developing an AI-powered surveillance tool designed to track real-time discussions of anti-China sentiments on social media platforms …

ABP News

Chinese Ai Tool Openai

OpenAI removes users in China, North Korea suspected of malicious activities

Trending 1 month, 2 weeks ago

OpenAI removes users in China, North Korea suspected of malicious activities

OpenAI has removed accounts of users from China and North Korea who the artificial intelligence company believes were using its technology for malicious purposes including surveillance and opinion-influence operations, the …

Deccan Chronicle

China North Korea Ai Openai