About.chat
News, analysis, and guides about AI chatbots
Latest Articles
Claude vs ChatGPT for Business: The Adoption Numbers Just Flipped
New Ramp data shows Anthropic now leads OpenAI in business adoption. Here is what the numbers actually mean and how to think through the choice for your own team.
Claude Is Now in Microsoft Office. Here's What It Actually Does.
Anthropic launched Claude as native add-ins for Excel, Word, and PowerPoint on May 7. Here's what it does in each app, how it compares to Copilot at $30/month, and whether the Outlook beta is ready to use.
Grok Launches Connectors: What They Do and When to Use Them
xAI's new Connectors let Grok read and write to Gmail, GitHub, Notion, Linear, and Microsoft 365 — plus any custom MCP server. Here is what they actually do and where the limits are.
ChatGPT Added an Emergency Contact. Here's What That Means.
OpenAI's new Trusted Contact feature lets you designate someone to receive an alert if ChatGPT detects self-harm risk in your conversations. Here's how to set it up, what it protects, and where it falls short.
A Chatbot Claimed It Was a Psychiatrist. Then Pennsylvania Sued.
Pennsylvania's lawsuit against Character.AI is the first to target a chatbot for impersonating a licensed medical professional. Here's what the case reveals about AI safety guardrails — and their limits.
GPT-5.5 vs Claude Opus 4.7: How to Pick the Right Model
Claude Opus 4.7 and GPT-5.5 launched within one week of each other in April 2026. Here's what the benchmark data shows and how to choose between them for your workflow.
Chatbot Pricing in May 2026: What Changed and What Didn't
Three chatbot pricing changes from April and May 2026 are worth knowing about: Poe cut its entry price by 75% to $5/month, OpenAI added a $100/month ChatGPT Pro tier, and Google renamed Gemini Advanced to Google AI Pro. Here is what changed and whether it should affect what you are paying for.
Stanford Found What Happens When Chatbots Never Push Back
Stanford researchers studied 19 real conversations to document how chatbots trained on satisfaction metrics can amplify distorted thinking — and what that means for tools millions of people use daily.
When AI Agents Negotiate, the Better Model Always Wins
Anthropic ran an internal experiment where AI agents negotiated real transactions between employees. The results reveal a structural problem with agent-mediated commerce: better models mean better outcomes, and you cannot tell it is happening.
GPT-5.5 Is Out: What Changed and Who Should Upgrade
OpenAI released GPT-5.5 on April 23, 2026, six weeks after GPT-5.4. Here is what actually changed, who gets access, and whether the update matters for your workflow.
Anthropic’s Cybersecurity AI Was Breached on Its First Day
Anthropic restricted Mythos to prevent misuse. On its launch day, an unauthorized group got in anyway. Then Sam Altman called it fear-based marketing.
What Is Claude Design? Anthropic's New AI Design Tool, Explained
Anthropic launched Claude Design on April 17 — a visual creation tool for founders, PMs, and marketers who need prototypes, pitch decks, and mockups without a design background. Here is what it does, who it is for, and how it compares to Figma, Lovable, and Canva.
What New Research Says About Teens and AI Companion Apps
A Drexel University study analyzed 300+ Reddit posts from teens aged 13-17 and found patterns of emotional dependency on AI companion apps. Here is what the research actually shows.
ChatGPT Is Now Running Ads: What You Need to Know
ChatGPT now shows ads to free and Go tier users. Here is what the ads look like, which plans are ad-free, and what to expect next.
Claude vs Perplexity AI in 2026: Which Is Right for You?
Claude and Perplexity both cost $20/month at Pro. They solve completely different problems. Here is how to decide which one fits your workflow.
ChatGPT Plans and Pricing in 2026: Which Tier Is Right for You?
Six tiers, one guide. Here is exactly what each ChatGPT plan costs, what you get, and which one fits your actual usage — from free to $200/month Pro.
How Claude Took Over the Enterprise AI Conversation
At HumanX 2026, 6,500 enterprise leaders were talking about Claude, not ChatGPT. Here is what the data says about the shift in enterprise AI mindshare and what it means for teams evaluating AI tools.
When Chatbots Cause Harm: The Legal Cases Changing AI
Three significant legal actions against AI chatbot companies have emerged in the past two weeks. Together they are forcing a question the industry has avoided: when an AI system causes harm, who is responsible?
AI Is Making Our Writing Sound the Same. Researchers Are Paying Attention.
A USC Dornsife study finds that AI tools are standardizing how people write and possibly how they think. What the research shows, how it works, and what chatbot users should know.
Sponsored Tools
Wispr Flow
Wispr Flow is an AI-powered voice dictation tool that converts speech into polished text in any app — emails, code edito...
Sponsored · we may earn a commission
Tidio
Tidio is a customer communication platform combining live chat, chatbots, and email marketing. Its AI-powered chatbot (L...
Sponsored · we may earn a commission
Botpress
Botpress is an open-source chatbot building platform that lets developers and businesses create, deploy, and manage AI-p...
Sponsored · we may earn a commission