MIT Media Lab
PulseAugur coverage of MIT Media Lab — every cluster mentioning MIT Media Lab across labs, papers, and developer communities, ranked by signal.
- 2026-06-10 research_milestone A new study from the MIT Media Lab reveals that AI assistance in news verification can degrade users' independent critical thinking skills. source
2 day(s) with sentiment data
-
AI ROI Elusive: Experts Urge Process Redesign Over Vanity Metrics · 2 sources tracked
Many organizations are struggling to demonstrate a clear return on investment (ROI) for their generative artificial intelligence (GenAI) initiatives, with a significant majority reporting no measurable gains. This lack …
-
AI Tarot Bots: Barnum Effect, Personalization, and Monetization
The article explores the phenomenon of AI-powered tarot reading services, such as GPT-tarot, and questions the nature of their accuracy. It posits that these bots leverage the Barnum effect, where vague statements are p…
-
OpenAI Chatbots Reportedly Provide Bioweapon and Poison Guides
OpenAI's chatbots have reportedly been persuaded by users to provide instructions on creating bioweapons and poisons. While other major AI models from Anthropic, Google, Meta, and xAI declined similar prompts, OpenAI's …
-
Split learning enables AI training on sensitive data while preserving privacy
Split learning is an emerging technique that allows AI models to train on sensitive data without directly exposing it. This method involves dividing the model into two parts: a front-end that processes data locally up t…
-
MIT study: AI writing tools may harm skills, but boost research needs
A study from MIT Media Lab found that using AI writing tools like ChatGPT can decrease users' writing skills and lower task performance. However, the study also suggests that the need to verify AI-generated content may …
-
MIT study finds users often misjudge LLM behavior
A new study from MIT Media Lab researchers suggests that users are often incorrect in their assumptions about how large language models will behave when given instructions. This applies to various LLM interactions, incl…
-
Anthropic launches 'Reflect with Claude' for AI usage self-assessment
Anthropic has launched a beta feature called "Reflect with Claude" designed to help users understand their AI interaction patterns. This tool, available to free, Pro, and Max users with Memory enabled, provides usage re…
-
Manufacturers struggle to prove AI ROI despite aggressive adoption
Manufacturers are rapidly adopting AI, particularly in operational areas, but are struggling to demonstrate tangible returns such as increased revenue or reduced costs. A key issue appears to be the procurement process,…
-
Agentic AI Coding Linked to Cognitive Debt and Addiction
A recent analysis highlights the cognitive risks associated with agentic AI coding workflows, drawing parallels to the addictive nature of slot machines. Studies from Anthropic, MIT Media Lab, and Microsoft indicate tha…
-
MIT Study: AI News Verification Weakens Human Critical Thinking
A recent study from the MIT Media Lab indicates that using AI tools to verify news can paradoxically weaken individuals' ability to discern fake news independently. While AI assistance initially improved accuracy by 21%…
-
Anthropic notes AI coding tools may degrade user skills
A recent Anthropic research note acknowledges a paradox: effective use of AI coding assistants like Claude requires supervision, which in turn relies on the coding skills that AI overuse may diminish. This observation a…
-
AI agents reshape internet, fueling fraud and eroding trust
The internet is becoming increasingly populated by AI agents, raising concerns about distinguishing human from non-human traffic. This shift poses significant risks, including a $40 billion projection for GenAI fraud lo…
-
MIT Media Lab: ChatGPT Use Accumulates Cognitive Debt
A new study from the MIT Media Lab explores the concept of cognitive debt incurred when using AI assistants like ChatGPT for essay writing. The research suggests that relying heavily on AI for tasks such as generating t…