🚀 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

Researchers have created an AI that can simulate your personality with 85% accuracy after a 2-hour chat!

Business Insider

Business Insider · 7m ago

Researchers have created an AI that can simulate your personality with 85% accuracy after a 2-hour chat!

Researchers from Google DeepMind have developed an AI system that can create highly accurate simulations of individual personalities. By analyzing speech patterns and decision-making tendencies, the AI constructs a digital personality that can answer questions with an impressive 85% accuracy compared to the real person. The researchers envision this technology as a game-changer for sociology, allowing for faster, cheaper, and more precise studies of human behavior. However, ethical considerations such as consent and potential misuse raise concerns. The technology could also have broader applications in personal AI assistants and human-robot interaction.

Related News

New AI model Life2vec may predict human lifespan: researchers

Economic Times

Economic Times · 1y ago

New AI model Life2vec may predict human lifespan: researchers

Researchers have developed an AI tool called Life2vec that can predict an individual's personality and even their lifespan using sequences of life events. The tool, trained on data from the entire population of Denmark, can surpass existing models in accuracy. However, the researchers caution that it should not be used for real-time predictions on real people. The goal is to use the tool as a foundation for future work in AI development, while taking a human-centered approach and not losing sight of the individuals in the massive data set.

Deci Open-Sources Foundational Models, Supasses Meta’s LLaMA & Stable Diffusion

Startup News FYI

Startup News FYI · 1y ago

Deci Open-Sources Foundational Models, Supasses Meta’s LLaMA & Stable Diffusion

Deci, an Israeli deep learning company, has launched generative AI foundation models that outperform Meta's algorithm by 15 times. These models, including DeciDiffusion and DeciLM 6B, are designed for efficient and cost-effective scaling of generative AI applications. Deci's open-source models and Infery LLM can reduce inference compute costs by up to 80%, making them accessible for AI researchers. These models cover various applications, including content and code generation, image creation, and chat applications, offering both speed and accuracy. They were developed using Deci's proprietary Neural Architecture Search (AutoNAC) technology.

China's Alibaba, South Korea's Naver launch GPT rival AI models

Economic Times

Economic Times · 1y ago

China's Alibaba, South Korea's Naver launch GPT rival AI models

Alibaba has introduced two new AI models, Qwen-VL and Qwen-VL-Chat, designed for image understanding and complex conversations. These models, based on Alibaba's large language model Tongyi Qianwen, are open source for global researchers. Qwen-VL can answer open-ended queries related to images and create captions, while Qwen-VL-Chat handles more intricate interactions, such as story writing and image creation based on user-input photos. Meanwhile, South Korean company Naver has launched its AI model HyperClovaX, accompanied by a generative AI search tool named Cue and a chat app called Clova X.

New AI model can hallucinate a game of 1993’s Doom in real time

Arstechnica

Arstechnica · 10m ago

New AI model can hallucinate a game of 1993’s Doom in real time

Google and Tel Aviv University researchers have developed GameNGen, an AI model that can simulate the 1993 game Doom in real-time using AI image generation techniques. The model, which functions as a limited game engine, has the potential to enable real-time video game synthesis by using AI to generate graphics on the spot. GameNGen can generate new frames of Doom gameplay at over 20 frames per second, and in tests, humans struggled to distinguish between the generated frames and actual gameplay footage. The model uses a modified version of Stable Diffusion 1.4 for image synthesis.

Hackers can read private AI assistant chats even though they’re encrypted

Arstechnica

Arstechnica · 1y ago

Hackers can read private AI assistant chats even though they’re encrypted

Researchers have discovered a technique that allows an attacker to decipher responses from AI assistants with high accuracy. By exploiting a side channel found in major AI assistants (excluding Google Gemini), the attack can infer the specific topic of 55% of all captured responses, often with precise wording. This passive attack can be carried out by someone monitoring the data packets between the assistant and user. The method involves monitoring token-length sequences, which are transmitted in real-time during a conversation. The researchers highlight the flawed encryption used by OpenAI and other chatbot providers, leaving private chats vulnerable to interception.

Here Come the AI Worms

Wired · 1y ago

Here Come the AI Worms

Researchers have created one of the first generative AI worms, highlighting the potential risks of connected, autonomous AI ecosystems. The worm, named Morris II, can spread from one system to another, potentially stealing data or deploying malware. By exploiting generative AI email assistants, the worm can steal data from emails and send spam messages. The researchers warn that as AI systems become more advanced and capable of completing tasks, developers and companies should be concerned about the security risks posed by generative AI worms. Mitigations include proper secure application design and monitoring, as well as keeping human approval in the loop.

OpenAI’s Custom chatbots are leaking their secrets

Wired · 1y 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.

South Korean team develops 'Iron Man' robot that helps paraplegics walk

Economic Times

Economic Times · 6m ago

South Korean team develops 'Iron Man' robot that helps paraplegics walk

South Korean researchers from the Exoskeleton Laboratory at KAIST have developed a lightweight wearable robot called the WalkON Suit F1. The robot is designed to help paraplegic individuals walk, navigate obstacles, and climb stairs. The suit can be locked onto the user, even while they are sitting in a wheelchair, and allows them to stand up and move. The suit weighs 50 kg and is powered by 12 electronic motors that simulate human joint movements. It also has sensors and cameras to enhance balance and perception for the user. The suit recently won a gold medal in the exoskeleton category at the Cybathlon 2024 event.

If you want to build solid business, don't give 100% daily: Ghazal

Inshorts · 4m ago

If you want to build solid business, don't give 100% daily: Ghazal

Mamaearth Co-founder Ghazal Alagh shared her "85% rule" to building business. "If you're serious about building a solid business, don't give your 100% every single day," she wrote, adding, "Over the years, working around the clock, I've realised that constantly pushing for a perfect 100% can actually backfire." Ghazal said one should instead learn focused effort and strategic prioritising.

Not working 70 hrs a week may not get you biggest house or car: Billionaire Khosla

Twitter · 1y ago

Not working 70 hrs a week may not get you biggest house or car: Billionaire Khosla

Billionaire venture capitalist Vinod Khosla said, "Not working 70 hrs/wk may not get you the biggest house or car to show to your neighbours, but you can make that choice." He was responding to an X user's question on NR Narayana Murthy suggesting 70-hour work week. Khosla added, "People who 'felt attacked' by [Murthy's suggestion] need mental health therapy."

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

June 2025

Helping your parents when you are miles away

The Pit Stop Your Cravings Deserve

The next generation E-commerce platform

Enter Ongoing Startup Showcase

Top Users

Trending News on Medial

Meta acquires voice startup Play ...

Exclusive: Porter turns profitabl ...

Who is Varun Mohan, the Windsurf ...

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