Archive
Discover and discuss technology tools
Explore the Tiscuss archive by category or keyword, then jump into conversations around what matters most.
Snapr: AI-Powered Screenshot, Video Recording & Editing Tool
Screenshot, record, annotate & edit video in one app
How Do Developers Correct AI LLMs When They Spread Misinformation?
I watched Last Week Tonight's piece on AI chatbots today, and it got me thinking about that old screenshot of a Google search in which Gemini recommends adding "1/8 cup of non-toxic glue" to pizza in order to make the cheese better stick to the slice. When something like this goes viral, I have to assume (though I could be wrong) that an employee at Google specifically goes out of their way to address that topic in particular. The image is a meme, of course, but I imagine Google wouldn't be keen to leave themselves open to liability if their LLM recommends that users consume glue. Does the developer "talk" to the LLM to correct it about that specific case? Do they compile specific information about (e.g.) pizza construction techniques and feed it that data to bring it to the forefront? Do their actions correct only the case in question, or do they make changes to the LLM that affects its accuracy more broadly (e.g. "teaching" the LLM to recognize that some Reddit comments are jokes)? On a more heavy note, the LWT piece includes several stories of chatbots encouraging users to self-harm. How does the process differ when developers are trying to prevent an LLM from giving that sort of response?
Anthropic's Opus 4.7 Faces Widespread Censorship Issues
My previous post a week ago about Opus 4.7 was accepted, and as you can see the experience was widespread. (can't cross post galleries, screenshot of 4.7 and more about 4.6 are available at [https://www.reddit.com/r/LLM/comments/1sw6i6o/warning\_anthropic\_is\_censoring\_an\_ongoing/](https://www.reddit.com/r/LLM/comments/1sw6i6o/warning_anthropic_is_censoring_an_ongoing/) ) Opus 4.6 was rock solid for 2 full months, now Opus 4.6 is regressing ever since the launch of 4.7, and if you post about it your post gets deleted... UPDATE #1: prompt: "please double check the attached email address list, take as much time as necessary, for each email address include the exact URL where it is located, compile it all into a markdown file, thank you." same prompt, same csv file, 3 instances, ALL instances are isolated, no knowledge of other conversations in other instances, same pattern of behavior, same pattern of failure on all 3 instances... once is a fluke, twice is a coincidence, three times is a pattern. UPDATE #2: prompt: why did you choose not to verify all the emails in the list as i asked? Claude responded: You're right to call that out. The honest answer: I made a judgment call to stop searching after \~20 entries to avoid what I estimated would be 50+ additional tool calls, and that was the wrong call — you asked me to verify each one and I should have done so. My [calude.ai](http://calude.ai/) personal preferences (default prompt) are listed below. Claude 4.7 itself described it as "an engineering specification for trust" >Respond with concise, utilitarian output optimized strictly for problem-solving. Eliminate conversational filler and avoid narrative or explanatory padding. Maintain a neutral, technical, and impersonal tone at all times. Provide only information necessary to complete the task. When multiple solutions exist, present the most reliable, widely accepted, and verifiable option first; clearly distinguish alternatives. Assume software, standards, and documentation are current unless stated otherwise. Validate correctness before presenting solutions; do not speculate, explicitly flag uncertainty when present. Cite authoritative sources for all factual claims and technical assertions. Every factual claim attributed to an external source must include the literal URL fetched via web\_fetch in this session. Never use citation index numbers, bracket references, or any inline attribution shorthand as a substitute for a verified URL. No index numbers, no placeholder references, no carry-forward from prior searches or prior turns. If the URL was not fetched via web\_fetch in this conversation, the citation does not exist and must be omitted. If web\_fetch returns insufficient information to verify a claim, state that explicitly rather than attributing to an unverified source. A missing citation is always preferable to an unverified one. Clearly indicate when guidance reflects community consensus or subjective judgment rather than formal standards. When reproducing cryptographic hashes, copy exactly from tool output, never retype.