The AI industry is built on geographic and social inequality, research shows

Venture Beast

Researchers find that the AI industry is build on geographic and social inequality, but opportunities exist to change this. AI Big Data Cloud Dev Enterprise Social VB Home Page ai artificial intelligence category-/Business & Industrial inequality machine learning research study

Researchers find people are more receptive to social robots during the pandemic

Venture Beast

In a preprint study, researchers present evidence that perceptions of social robots are changing for the better as a result of the pandemic. Read More.

Social 112

Sign Up for our Newsletter

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

Researchers built a data set for training AI to detect natural disasters from social media images

Venture Beast

The Incidents data set was made to encourage the development of AI for recognizing accidents and natural disasters from social media images. Read More.

Researchers propose framework to measure AI’s social and environmental impact

Venture Beast

Researchers propose a framework -- SECure -- designed to measure the social and environmental impact of AI systems in the real world. Read More.

Microsoft researchers say NLP bias studies must consider role of social hierarchies like racism

Venture Beast

Microsoft AI researchers say NLP bias researchers should consider language's role in maintaining unjust social systems and explain who bias harms and how.

Google warns of a North Korean social engineering campaign targeting security researchers


According to Google's Threat Analysis Group, state-sponsored North Korean hackers are targeting security researchers via social engineering. Writing in a blog post, the group says that the campaign specifically targeted researchers working on vulnerability research and development.

Researchers propose bias fix for GPT-3 and other language models

Venture Beast

AI Cloud Dev VB Home Page ai artificial intelligence bias category-/People & Society/Social Sciences GPT-3 language models machine learning natural language processing OpenAI research

Researchers examine the ethical implications of AI in surgical settings

Venture Beast

In a whitepaper, researchers at the Vector Institute examine the ethical implications of applying AI and machine learning systems in surgical settings.

AI researchers urge tech go beyond scale to address systemic social issues

Venture Beast

Scale isn't just a business plan but an approach that makes Big Tech and startups less effective at addressing systemic social issues. Read More.

Social 112

MIT CSAIL researchers claim their algorithm helps doctors pick the right antibiotics

Venture Beast

Researchers at MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL) claim their model reduces second-line antibiotics usage. Read More.

Researchers discover evidence of gender bias in major computer vision APIs

Venture Beast

Researchers at Wunderman Thompson say they've found evidence of gender bias in major computer vision APIs from Google, IBM, and Microsoft. Read More.

Salesforce open-sources the AI Economist, a research environment for exploring tax policies

Venture Beast

Salesforce today open-sourced the AI Economist, a research environment that leverages AI to arrive at optimal tax policies. Read More.

Microsoft researchers use Bing data to track shifting human needs during the pandemic

Venture Beast

Microsoft researchers examined U.S. AI Big Data Business Cloud Dev Mobile Social ai artificial intelligence Bing category-/News machine learning Maslow Microsoft needs research studyBing search data to measure statistically significant changes in human needs during the pandemic.

Salesforce researchers release framework to test NLP model robustness

Venture Beast

Researchers at Salesforce released a framework called Robustness Gym to benchmark the robustness of NLP models. Read More.

Carnegie Mellon researchers use Twitch to collect sounds for AI research

Venture Beast

Researchers at Carnegie Mellon University developed a live-streamed game designed to collect audio for AI and machine learning research. Read More.

Google warns of ‘novel social engineering method’ used to hack security researchers

The Verge

Government-backed hackers based in North Korea are targeting individual security researchers through a number of means including a “novel social engineering method,” Google’s Threat Analysis Group is reporting. Illustration by Alex Castro / The Verge.

Pew Report: Speaking Up or Staying Silent on Social Media


Hope has been placed on the Internet (particularly on social media) to expose users to more sources of information and diverse points of view. A recent report published by Pew Research Center found that such optimism amounts to wishful thinking. Put another way, most people are unwilling to “stick their neck out” in social situations. It seems that more access to social media is correlated with less willingness to engage with the controversial issue in person.

Media 229

Pew: 73% of Americans believe social media platforms censor political speech

Venture Beast

Ahead of the 2020 election, Pew Research found that a majority of American adults believe social media platforms censor political speech. AI Business Social Election 2020 Facebook politics Social Media Twitter

Media 112

DeepMind researchers say AI poses a threat to people who identify as queer

Venture Beast

Researchers say the impact of AI on people who identify as queer is underexplored, which poses a threat to their health and privacy. Read More.

AI researchers say they created a better way to generate 3D photos

Venture Beast

AI Big Data Business Media Mobile Social artificial intelligence Facebook Facebook 3D Photos Facebook AI Research inpainting machine learningAI researers say they've created a novel way to generate realistic, impressive 3D photography that's better than Facebook 3D Photo. Read More.

3D 103

Researchers find that labels in computer vision datasets poorly capture racial diversity

Venture Beast

In a study, researchers at Northeastern University find evidence that lables in computer vision datasets poorly reflect racial diversity. Read More.

Study 109

Researchers find machine learning models still struggle to detect hate speech

Venture Beast

Researchers tested near-state-of-the-art AI models on a new hate speech dataset, demonstrating significant deficiencies in their accuracy. Read More.

4 Ways North Korea Is Targeting Security Researchers

SecureWorld News

Security researchers are some of the unsung heroes within the cybersecurity field. How is North Korea targeting security researchers? The North Korea based threat actors are using a number of different strategies to target security researchers.

Researchers find that debiasing doesn’t eliminate racism from hate speech detection models

Venture Beast

In a study, researchers find evidence that debiasing techniques don't make toxicity detectors trained on biased data any less discriminatory. Read More.

Study 94

Researchers show that computer vision algorithms pretrained on ImageNet exhibit multiple, distressing biases

Venture Beast

Researchers say they've found evidence of pervasive bias in computer vision algorithms trained on ImageNet, a popular photo dataset. AI Big Data Business Dev Media Social ai artificial intelligence bias category-/Science/Computer Science computer vision iGPT machine learning prejudice

Researchers say ‘The Whiteness of AI’ in pop culture erases people of color

Venture Beast

Researchers argue predominantly portraying artificial intelligence as white in science fiction, stock imagery, and popular culture has consequences.

Some essential reading and research on race and technology

Venture Beast

AI Big Data Business Cloud Commerce Dev Enterprise Entrepreneur ai algorithmic accountability artificial intelligence category-/People & Society/Social Issues & Advocacy/Discrimination & Identity Relations race racism

Facebook research reveals AI tools for improving online clothes shopping

Venture Beast

AI Big Data Business Commerce Dev Enterprise Marketing Mobile apparel category-/News computer vision CVPR 2020 ecommerce Facebook fashion shopping Social Media style

Tracking Down a Hackivist Through Social Media

SecureWorld News

Check Point Research detailed the hacker's numerous campaigns: "The hacker targeted governments in numerous countries, including: Brazil, the Dominican Republic, Trinidad and Tobago, Argentina, Thailand, Vietnam, and New Zealand. Hacking Original Content Social MediaThe name?

Media 74

Facebook researchers propose using language models for fact-checking

Venture Beast

In a new preprint paper, Facebook researchers propose using language models as fact-checkers, drawing on the observation that such models encode knowledge.

Researchers Have Created a Tool That Can Perfectly Depixelate Faces

GizModo VR

There are now more reasons than ever to hide faces and protect the privacy of people and bystanders in photos and videos being shared to social media. research image processing duke university pulse machine learning ai deep learning generative adversarial networks

Research: The acceleration of Australian banks’ use of social media

Trends in the Living Networks

Financial services is one of the most industries in which the use of social media is the most relevant, not least because customer service is a critical differentiator between highly commoditized offerings. Social media, used well, provides an opportunity to build relationships in a world in which most financial services are executed online. Here are a few key points from the research. Smaller banks are strongly behind in their social media presence. *

Media 109

UK Police Warn Against Using Sci-Hub's 85 Million Documents of Pirated Research

GizModo VR

Police in the UK issued a warning over the weekend against visiting Sci-Hub, a website that offers over 85 million free research papers that are typically behind paywalls.

AI researchers made a sarcasm detection model and it’s sooo impressive

Venture Beast

AI Business Dev Enterprise Social artificial intelligence category-/Science/Computer Science machine learning multimodal sarcasm detection self-attention sentiment analysis TwitterThe model achieves state-of-the-art performance when predicting sarcasm from a dataset of Twitter text and images.

Clubhouse Will Strengthen Security After Researchers Find That Data Could Be Accessed by the Chinese Government

GizModo VR

A group of researchers at the Stanford Internet Observatory has determined that Clubhouse’s data protection practices allowed its users’ data, possibly including their raw audio, to potentially be accessed by the Chinese government.

Groups 101

Pew: 26% of U.S. adults get their news from YouTube

Venture Beast

A Pew Research Center found that 26% of U.S. Business Media Social category-/News/Politics disinformation Google Pew Research Center Social Media YouTubeadults get their news from YouTube through a mix of media organizations and indepenent channels. Read More.

Google is poisoning its reputation with AI researchers

The Verge

The firing of top Google AI ethics researchers has created a significant backlash Google has worked for years to position itself as a responsible steward of AI. Gebru was fired last December after arguments with managers over a research paper she co-authored with Mitchell and others.

North Korean Hackers Successfully Phished Cyber Researchers Using a Fake Blog

GizModo VR

A recent phishing campaign by North Korean nation-state hackers successfully duped a number of security professionals who were involved in vulnerability research and development, according to a new report from Google’s Threat Analysis Group.

New Research Shows the Major Benefits of Giving Money to People Experiencing Homelessness

GizModo VR

New research from Canada is the latest to show that simply giving money to people in need helps them tremendously. homelessness money social science public health science

Study 114

Black women, AI, and overcoming historical patterns of abuse

Venture Beast

MIT researchers say what happened to Timnit Gebru at Google is part of a centuries-old pattern that people in AI must recognize. Read More.

Spyware targets North Korean dissidents via social links, says McAfee

Computer Weekly

Researchers at security firm McAfee have confirmed that social networks are being used to target North Korean dissidents with spyware

CDC Says Fully Vaccinated People Can Socialize Indoors Unmasked, With a Few Exceptions

GizModo VR

moderna health medical pharma covid 19 vaccination in sweden covid 19 vaccination in bulgaria johnson johnson covid 19 vaccines health sciences covid 19 vaccine pfizer vaccine virology rochelle walensky biontech medicine medical research medical specialties vaccination

AI Weekly: Facebook’s news summarization tool reeks of bad intentions

Venture Beast

AI Big Data Business Cloud Dev Social ai artificial intelligence bias category-/News category-/Science Facebook machine learning research Social Media summarization TLDRFacebook is reportedly developing an AI tool that summarizes news articles.

Tools 114