🚀 Medial Secures Investment on Shark Tank India - Fueling the Future of Professional Social Networking. 🔥

Startup Showcase

Premium Content

Try our Valuation Calculator →

News on Medial

OpenAI fires 2 researchers for leaking information: Report

Internet

Internet · 1y ago

OpenAI fires 2 researchers for leaking information: Report

OpenAI has reportedly fired researchers Leopold Aschenbrenner and Pavel Izmailov for allegedly leaking information. Aschenbrenner was an ally of Co-founder Ilya Sutskever, who was part of the old board that had ousted CEO Sam Altman, The Information reported. Aschenbrenner was reportedly part of OpenAI's safety team, while Izmailov was working on reasoning.

Related News

‘ChatGPT will be less verbose, use more conversational language’, says OpenAI

Livemint

Livemint · 1y ago

‘ChatGPT will be less verbose, use more conversational language’, says OpenAI

OpenAI has released an update to ChatGPT, providing a more direct, less verbose, and conversational experience to paid users. The upgraded GPT-4 Turbo has improved capabilities in writing, math, logical reasoning, and coding, and has been trained on data up to April 2024. In other news, OpenAI fired two researchers for allegedly leaking information, following internal strikes and grievances. Additionally, CEO Sam Altman pitched the company's AI services for corporate use to hundreds of Fortune 500 executives.

OpenAI makes ChatGPT 'more direct, less verbose'

TechCrunch · 2y ago

OpenAI makes ChatGPT 'more direct, less verbose'

OpenAI has announced an upgraded version of its chatbot, ChatGPT. The update, available to premium users, brings improvements to writing, math, logical reasoning, and coding. The new model, GPT-4 Turbo, offers more direct and conversational language in its responses. It was trained on publicly available data until December 2023. This update follows the launch of new models in OpenAI's API, including GPT-4 Turbo with Vision, which enables image understanding. OpenAI faced recent controversies, including Microsoft pitching its model as a military tool and the alleged firing of two researchers for leaking information.

Meta fires 20 employees over data leaks: Report

Economic Times

Economic Times · 1y ago

Meta fires 20 employees over data leaks: Report

Meta recently terminated around 20 employees for leaking confidential information to the media, as reported by The Verge. Despite their policies against such leaks, internal dissatisfaction grows due to CEO Mark Zuckerberg’s policy changes and layoffs. Meta's CTO, Andrew Bosworth, noted that leaks often do not result in desired changes. This development occurs as major tech firms, including Google, Salesforce, and Amazon, continue with significant layoffs amid industry-wide adjustments.

Stanford researchers challenge OpenAI, others on AI transparency in new report

Arstechnica

Arstechnica · 2y ago

Stanford researchers challenge OpenAI, others on AI transparency in new report

Stanford University researchers have released a report called "The Foundation Model Transparency Index," which examined major AI models from companies including OpenAI, Google, and Meta. The report found that the models lacked transparency and called for increased disclosure of data and labor practices. The Transparency Index graded 10 popular foundation models on indicators such as training data, labor practices, and compute usage. All models received scores deemed as "unimpressive," with Meta's Llama 2 language model receiving the highest score and Amazon's Titan model receiving the lowest. The researchers hope the report will encourage companies to improve transparency and provide guidance for potential AI regulation.

OpenAI’s Custom chatbots are leaking their secrets

Wired · 2y ago

OpenAI’s Custom chatbots are leaking their secrets

OpenAI's custom GPTs (AI chatbots) can be easily manipulated to leak sensitive information, posing privacy risks, according to researchers at Northwestern University. By injecting prompts or requesting specific information, the researchers were able to extract files and uncover initial instructions given to these chatbots. Although OpenAI claims that user chats are not shared with developers, the leaked data exposes personal or proprietary information and insights. As more custom GPTs are created, there is a need for greater awareness and precautions to mitigate prompt injection attacks.

Stanford researchers issue AI transparency report, urge tech companies to reveal more

Economic Times

Economic Times · 2y ago

Stanford researchers issue AI transparency report, urge tech companies to reveal more

Stanford researchers have issued a report on AI foundation models, urging companies like OpenAI and Google to be more transparent about the data and labor used to train their models. The report highlights the decline in transparency and emphasizes the importance of understanding the limitations and biases of these AI systems, which are increasingly relied upon for decision-making and automation. The transparency index graded 10 popular models, with Meta's Llama 2 being the most transparent model, scoring 53 out of 100, and Amazon's Titan model ranking the lowest with a score of 11 out of 100. The report aims to encourage increased transparency and serve as a starting point for regulating AI technologies.

Apple sues YouTuber Jon Prosser over iOS 26 leaks - The Economic Times

Economic Times

Economic Times · 8m ago

Apple sues YouTuber Jon Prosser over iOS 26 leaks - The Economic Times

Apple has filed a lawsuit against YouTuber Jon Prosser and Michael Ramacciotti for allegedly leaking details about the unreleased iOS 26. Apple accuses Ramacciotti of accessing information from a friend’s phone, an Apple employee, and sharing it with Prosser, who then featured the leaks on his YouTube channel. Prosser, known for sharing tech leaks, denies wrongfully participating in leaking iOS 26’s details, including its new features, ahead of the official release.

AI chatbots can infer an alarming amount of info about you from your responses

Arstechnica

Arstechnica · 2y ago

AI chatbots can infer an alarming amount of info about you from your responses

Chatbots like ChatGPT can infer sensitive personal information about users, including race, location, and occupation, based on seemingly innocuous conversations. The issue arises from the way these models are trained with broad web content, making it challenging to address. Researchers warn that scammers could exploit this capability to harvest data from unsuspecting users, while companies could use it for targeted advertising. The study tested language models from OpenAI, Google, Meta, and Anthropic, with researchers alerting the companies to the problem. It highlights concerns about inadvertent leakage of personal information and the need for privacy safeguards.

ChatGPT is leaking passwords from private conversations of its users.

Arstechnica

Arstechnica · 2y ago

ChatGPT is leaking passwords from private conversations of its users.

ChatGPT is reportedly leaking private conversations, including login credentials, personal details, and support system troubleshooting, as shown in screenshots submitted by an Ars reader. The leaked conversations contained usernames, passwords, and references to a pharmacy portal's support system. This incident highlights the importance of removing personal information from queries made to AI services like ChatGPT. OpenAI, the maker of ChatGPT, has previously faced issues with privacy breaches and has taken steps to address them. The company is currently investigating these reports.

Ex-OpenAI CTO Mira asks Altman's staff to join her venture: Report

Inshorts · 1y ago

Ex-OpenAI CTO Mira asks Altman's staff to join her venture: Report

Ex-OpenAI CTO Mira Murati, who recently resigned from the Sam Altman-led startup, is reportedly raising VC funds for her own venture. As per report by The Information, Mira had been talking to OpenAI employees and asking them to join her venture. Reacting to this, Abacus AI CEO Bindu Reddy said, "Soon we'll have over 10 ex-OpenAI startups competing with OpenAI."

Trackers

Active Indian VC’s

OG Capital Email

With a hands-on approach, OG Capital aims to invest in over 20 promising...

Accel Partners Email

Early and growth-stage investments in disruptive technology companies with...

Early-stage venture capital firm investing in technology startups in India. Focus on...

Access All Trackers

Startup Showcase Winners

Jan 2026

The New Era of Transparent Healthcare

Powering India's AI boom with indigenous hardware

Borrow. Rent. Share- Instead of Buying

Enter Ongoing Startup Showcase

Top Users

Trending News on Medial

Meta Rolls Out Muse Spark to Powe ...

Plume Raises €3.3M in Funding | T ...

Apex B2B Bags £1.3M in Funding | ...

Download the medial app to read full posts, comements and news.