\n<\/p>\n
Stone Crab Claws Ready to Eat<\/span><\/p>\n getty<\/small><\/div>\n<\/div>\n<\/figure>\n It\u2019s the most famous AI assistant around, right now \u2013 but OpenClaw remains seriously controversial. Some users who have tried it tout it abilities, while others insist that security issues make it a net negative, for handling sensitive assets like an inbox.<\/p>\n Let\u2019s face it, the technology is new, and relatively untested. The more savvy among us, unwilling to act as digital guinea pigs, set up a less sensitive sandbox for OpenClaw to claw around in, and see what shakes out, in a low-stakes environment. Those who actually fork over their passwords sometimes see catastrophic results.<\/p>\n You can read the reporting from The Information, titled: OpenClaw Struggles to Grow Up After Overnight Success, by Juro Osawa and Rocket Drew, to get a better picture of this, or alternately, talk to people who have taken the plunge, to get a better understanding of how OpenClaw works.<\/p>\n On the front end, in more of an administrative category, you have a range of issues related to coding or operational problems with OpenClaw. These are not going to result in pilfered assets, or inappropriate trashing of personal files, or anything like that. You\u2019ll see them pop up, in WSL or Windows, or whatever your environment is, and the result is going to be a hung session.<\/p>\n Specifically, an article by Kunpeng-AI at Dev<\/u><\/a> illustrates the error messages you\u2019ll see with a range of config errors: an invalid API key will return: \u201cError: Invalid API key,\u201d a JSON error will return: \u201cError: Invalid JSON in config file at line XX.\u201d A runtime error might mean lost messages or webhook verification fails.<\/p>\n Then there\u2019s that whole host of nightmare scenarios that don\u2019t have to do with operational snafus in the code. Instead, they represent a well-meaning OpenClaw doing something that just isn\u2019t aligned with user desire. If it was human, we would say the claw is \u201cmisguided\u201d and acting out of shoddy intent.<\/p>\n For example, so many who have given OpenClaw access have seen it go in and mass-delete emails for no readily apparent reason. <\/p>\n \u201cMeta\u2019s Director of AI and Safety Alignment wanted to clean up her inbox, so she set up an OpenClaw AI agent and told it to \u2018confirm before acting,\u2019\u201d writes Gil Feig at TechRadar<\/u><\/a>. \u201cBut it didn’t. Instead, the OpenClaw agent mass-deleted hundreds of emails while she scrambled to shut it down from another device.\u201d<\/p>\n When that happens to someone so close to the AI world, you wonder what\u2019s happening to everyone else.<\/p>\n Other accounts have to do with OpenClaw providing faulty information, or failing to safeguard process in some mission-critical way. Take a look at articles like this going into a range of problem scenarios.<\/p>\n When it comes to one of the biggest model makers in the industry, the realm of the Amodei brethren, there\u2019s been quite a bit of kerfuffle between Anthropic and OpenClaw, starting with the former\u2019s pushback on the original name, Clawdbot, and the banning of Peter Steinberger from the Claude platform recently. Part of this may be around the bleeding of token volume through third-party harnesses, which I wrote about last week, but there are other issues, too.<\/p>\n \u201cI’ve been working on getting the claude -p fallback feature working after Boris confirmed that it’s a classifier bug and not intentional,\u201d Steinberger said of the ban, which was quickly reversed, according to reporting by Aman Gupta at Live Mint<\/u><\/a>. \u201cWe’re still blocked, and it seems that got me banned, too.\u201d<\/p>\n It seems, from some light reading, that the -p flag deals with bad API calls. Anyway, it\u2019s an illustration of the tension between the two projects. Anthropic, it seems, isn\u2019t a fan of OpenClaw \u2013 and a lot of security pros aren\u2019t, either.<\/p>\nThe Glitching: Configuration Errors, Runtime Errors, and More<\/h2>\n
Running Amok<\/h2>\n
OpenClaw and Anthropic<\/h2>\n
Cumbersome Technology? <\/h2>\n