Author HubMaximilian Schreiner

Max is the managing editor of THE DECODER, bringing his background in philosophy to explore questions of consciousness and whether machines truly think or just pretend to.

AI research

Sep 17, 2025

OpenAI outperforms humans and Google at the world's top collegiate programming contest

AI in practice

Sep 15, 2025Sep 15, 2025

New data from OpenAI and Anthropic show how people actually use ChatGPT and Claude

AI in practice

Aug 7, 2025Aug 7, 2025

OpenAI launches GPT-5 as a unified system with adaptive reasoning for complex tasks

AI research

Jul 25, 2023Jul 25, 2023

STEVE-1 is a chatbot that plays Minecraft

AI in practice

Jul 25, 2023Jul 25, 2023

Update

OpenAI's ChatGPT app for Android is now available

AI in practice

Jul 22, 2023Jul 22, 2023

Two systems of thinking - when AI leverages human psychology

AI and society

Jul 21, 2023

Short

Dave Willner, head of trust and safety at OpenAI, has transitioned to an advisory role after a year and a half in the position. Willner, who previously led trust and safety teams at Facebook and Airbnb, played a significant part in discussions surrounding regulation and responsibility in the AI world after the launch of ChatGPT.

"Anyone with young children and a super intense job can relate to that tension, I think, and these past few months have really crystallized for me that I was going to have to prioritize one or the other," he said in a LinkedIn post.

His departure occurs at a critical time for AI, as OpenAI's president Greg Brockman is meeting with the White House alongside executives from major tech companies to discuss shared safety and transparency goals.

Reuters

AI in practice

Jul 19, 2023Jul 19, 2023

Study shows evidence of degradation in ChatGPT's performance since March

AI and society

Jul 18, 2023

Short

Surveillance company Facewatch is experiencing "exponential" demand for its AI-powered facial recognition system to identify repeat shoplifters, founder Simon Gordon said. The system works by having a store manager review security camera footage after an item is stolen, logging the thief's image into Facewatch's system for future alerts.

Critics say this technology infringes on privacy rights and often makes mistakes. But according to Gordon, the system currently has a 99.85% accuracy rate and data is only stored for two weeks, half the time of traditional security camera footage, reports CNN.

CNN

AI research

Jul 18, 2023Jul 18, 2023

GPT-4 beats nearly 2,700 students in creative thinking

AI research

Jul 16, 2023Jul 16, 2023

New method improves Stable Diffusion without training data

AI in practice

Jul 15, 2023Jul 15, 2023

Update

OpenAI DALL-E 2 Prompt Guide: How to use the generative AI model

AI in practice

Jul 15, 2023Jul 15, 2023

Short

Google's MatCha is a foundation model trained for both chart de-rendering and mathematical reasoning. Chart de-rendering explores the reverse engineering of charts, plots, or graphics to reveal their underlying data table or code, while math reasoning seeks to solve question-based problems on textual mathematical datasets. By combining these tasks, MatCha significantly outperforms existing models for visual language understanding of charts. The researchers also proposed DePlot, a model built on top of MatCha for improved reasoning on charts through translation to tables.

Google

Google News

Bank details