The search engine was impressive enough but I think this implementation was a nice cherry on top."},"story_title":{"matchLevel":"none","matchedWords":[],"value":"Launch HN: Exa (YC S21) \u2013 The web as a database"}},"_tags":["comment","author_euvin","story_43906841"],"author":"euvin","comment_text":"I found the hallucination detector demo: https://demo.exa.ai/hallucination-detector
The search engine was impressive enough but I think this implementation was a nice cherry on top.","created_at":"2025-05-07T01:14:54Z","created_at_i":1746580494,"objectID":"43911277","parent_id":43906841,"story_id":43906841,"story_title":"Launch HN: Exa (YC S21) \u2013 The web as a database","updated_at":"2025-05-07T01:34:34Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"tingfirst"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai","hallucination"],"value":"Catching AI Hallucination in SQL: The Chess Example"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["ai","hallucination","detect"],"value":"https://www.timeplus.com/post/ai-chess-hallucination-detection"}},"_tags":["story","author_tingfirst","story_44954304"],"author":"tingfirst","children":[44954305],"created_at":"2025-08-19T17:54:19Z","created_at_i":1755626059,"num_comments":0,"objectID":"44954304","points":3,"story_id":44954304,"title":"Catching AI Hallucination in SQL: The Chess Example","updated_at":"2026-03-05T22:31:29Z","url":"https://www.timeplus.com/post/ai-chess-hallucination-detection"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"rymc"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["hallucination","detect"],"value":"LLM Hallucination Detection Leaderboard"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["ai","hallucination","detect"],"value":"https://huggingface.co/spaces/kluster-ai/LLM-Hallucination-Detection-Leaderboard"}},"_tags":["story","author_rymc","story_44502627"],"author":"rymc","children":[44502650],"created_at":"2025-07-08T18:23:38Z","created_at_i":1751999018,"num_comments":0,"objectID":"44502627","points":1,"story_id":44502627,"title":"LLM Hallucination Detection Leaderboard","updated_at":"2025-07-08T18:27:59Z","url":"https://huggingface.co/spaces/kluster-ai/LLM-Hallucination-Detection-Leaderboard"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"skeptrune"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["hallucination","detect"],"value":"Accurate Hallucination Detection with NER"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["ai","hallucination","detect"],"value":"https://trieve.ai/accurate-hallucination-detection-with-ner"}},"_tags":["story","author_skeptrune","story_42629387"],"author":"skeptrune","created_at":"2025-01-08T00:16:45Z","created_at_i":1736295405,"num_comments":0,"objectID":"42629387","points":1,"story_id":42629387,"title":"Accurate Hallucination Detection with NER","updated_at":"2025-01-08T00:18:45Z","url":"https://trieve.ai/accurate-hallucination-detection-with-ner"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"jaquelinejaque"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["ai","hallucination","detect"],"value":"Show HN: I built an 11-LLM consensus engine to detect AI hallucination"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/jaquelinejaque/quorum-saas-starter"}},"_tags":["story","author_jaquelinejaque","story_48596771","show_hn"],"author":"jaquelinejaque","children":[48596816,48598208,48600392,48601796,48603734,48604303,48608885],"created_at":"2026-06-19T09:53:50Z","created_at_i":1781862830,"num_comments":5,"objectID":"48596771","points":6,"story_id":48596771,"title":"Show HN: I built an 11-LLM consensus engine to detect AI hallucination","updated_at":"2026-08-16T12:32:13Z","url":"https://github.com/jaquelinejaque/quorum-saas-starter"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"reboy"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["ai","hallucination","detect"],"value":"Show HN: Halgorithem \u2013 an open-source tool for detecting AI hallucinations"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/TangibleResearch/Halgorithem"}},"_tags":["story","author_reboy","story_48722856","show_hn"],"author":"reboy","children":[48722857],"created_at":"2026-06-29T18:07:33Z","created_at_i":1782756453,"num_comments":0,"objectID":"48722856","points":1,"story_id":48722856,"title":"Show HN: Halgorithem \u2013 an open-source tool for detecting AI hallucinations","updated_at":"2026-06-29T18:13:14Z","url":"https://github.com/TangibleResearch/Halgorithem"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"webmaven"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["ai","hallucination","detect"],"value":"Galileo takes on detecting AI hallucinations, but new metrics are needed"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["ai","hallucination","detect"],"value":"https://diginomica.com/galileo-takes-detecting-ai-hallucinations-new-metrics-are-needed"}},"_tags":["story","author_webmaven","story_38434301"],"author":"webmaven","created_at":"2023-11-27T16:29:17Z","created_at_i":1701102557,"num_comments":0,"objectID":"38434301","points":1,"story_id":38434301,"title":"Galileo takes on detecting AI hallucinations, but new metrics are needed","updated_at":"2024-09-20T15:47:26Z","url":"https://diginomica.com/galileo-takes-detecting-ai-hallucinations-new-metrics-are-needed"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"hirundo"},"comment_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["ai","hallucination","detect"],"value":"> Real Risk: Overreliance on inaccurate tools
We might be able to remediate this somewhat by designing AIs that are purposely below some threshold of reliability. Their current tendency to hallucinate will discourage over reliance if we can't or don't fix it.
It's a little like the distinctive smell that is added to propane to make it easier to detect leaks. By adding, or not removing, easily detectable hallucinations from AI, it's easier to detect that the source must be checked.
We desperately want oracles and will quickly latch on to highly unreliable sources at the drop of a hat, to judge by various religious, political and economic trends. It's inevitable that many will turn AIs into authorities rather than tools as soon as they can justify it. We could delay that by making it less justifiable."},"story_title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"A misleading open letter about sci-fi AI dangers ignores the real risks"},"story_url":{"matchLevel":"none","matchedWords":[],"value":"https://aisnakeoil.substack.com/p/a-misleading-open-letter-about-sci"}},"_tags":["comment","author_hirundo","story_35373042"],"author":"hirundo","children":[35374536,35386272],"comment_text":"> Real Risk: Overreliance on inaccurate tools
We might be able to remediate this somewhat by designing AIs that are purposely below some threshold of reliability. Their current tendency to hallucinate will discourage over reliance if we can't or don't fix it.
It's a little like the distinctive smell that is added to propane to make it easier to detect leaks. By adding, or not removing, easily detectable hallucinations from AI, it's easier to detect that the source must be checked.
We desperately want oracles and will quickly latch on to highly unreliable sources at the drop of a hat, to judge by various religious, political and economic trends. It's inevitable that many will turn AIs into authorities rather than tools as soon as they can justify it. We could delay that by making it less justifiable.","created_at":"2023-03-30T15:27:36Z","created_at_i":1680190056,"objectID":"35374208","parent_id":35373042,"story_id":35373042,"story_title":"A misleading open letter about sci-fi AI dangers ignores the real risks","story_url":"https://aisnakeoil.substack.com/p/a-misleading-open-letter-about-sci","updated_at":"2024-09-20T13:41:31Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"MrSteaddy"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["ai","hallucination","detect"],"value":"InsAIts: Monitoring for AI-AI comms. Detect hallucinations before propagation"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/Nomadu27/InsAIts"}},"_tags":["story","author_MrSteaddy","story_46888113"],"author":"MrSteaddy","children":[46888114],"created_at":"2026-02-04T16:46:09Z","created_at_i":1770223569,"num_comments":2,"objectID":"46888113","points":1,"story_id":46888113,"title":"InsAIts: Monitoring for AI-AI comms. Detect hallucinations before propagation","updated_at":"2026-03-05T23:31:30Z","url":"https://github.com/Nomadu27/InsAIts"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"olivierc_RM"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["ai","hallucination","detect"],"value":"We built live AI evaluation to detect hallucinations, grounding issues, and drift as outputs are generated. This helps teams move GenAI from pilots into production."},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["ai","hallucination","detect"],"value":"Show HN: Live AI Evaluation to Detect Hallucinations in Real Time"},"url":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"https://ragmetrics.ai/live-ai-evaluation"}},"_tags":["story","author_olivierc_RM","story_46290592","show_hn"],"author":"olivierc_RM","created_at":"2025-12-16T16:30:52Z","created_at_i":1765902652,"num_comments":0,"objectID":"46290592","points":1,"story_id":46290592,"story_text":"We built live AI evaluation to detect hallucinations, grounding issues, and drift as outputs are generated. This helps teams move GenAI from pilots into production.","title":"Show HN: Live AI Evaluation to Detect Hallucinations in Real Time","updated_at":"2026-03-05T23:13:33Z","url":"https://ragmetrics.ai/live-ai-evaluation"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"matteo1782"},"story_text":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai","hallucination"],"value":"Phantom Guard is a CLI tool that catches "slopsquatting" attacks before they compromise your supply chain.
The attack vector: AI assistants hallucinate package names \u2192 attackers register those names with malware \u2192 developers install malware thinking it's legit.
How it works:\n1. Checks if packages exist on registries\n2. Matches against 10+ AI hallucination patterns\n3. Detects typosquats of top 3000 packages\n4. Analyzes metadata (age, downloads, maintainers)
```\npip install phantom-guard\nphantom-guard validate flask-gpt-helper\n# HIGH_RISK: Package not found, matches pattern\n```
Performance: <10ms cached, <200ms uncached.
Try the live demo: https://matte1782.github.io/phantom_guard/
GitHub: https://github.com/matte1782/phantom_guard"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai","detect"],"value":"Show HN: Phantom Guard \u2013 Detect AI-hallucinated package attacks"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/matte1782/phantom_guard"}},"_tags":["story","author_matteo1782","story_46478932","show_hn"],"author":"matteo1782","children":[46489003],"created_at":"2026-01-03T17:03:24Z","created_at_i":1767459804,"num_comments":2,"objectID":"46478932","points":2,"story_id":46478932,"story_text":"Phantom Guard is a CLI tool that catches "slopsquatting" attacks before they compromise your supply chain.
The attack vector: AI assistants hallucinate package names \u2192 attackers register those names with malware \u2192 developers install malware thinking it's legit.
How it works:\n1. Checks if packages exist on registries\n2. Matches against 10+ AI hallucination patterns\n3. Detects typosquats of top 3000 packages\n4. Analyzes metadata (age, downloads, maintainers)
```\npip install phantom-guard\nphantom-guard validate flask-gpt-helper\n# HIGH_RISK: Package not found, matches pattern\n```
Performance: <10ms cached, <200ms uncached.
Try the live demo: https://matte1782.github.io/phantom_guard/
GitHub: https://github.com/matte1782/phantom_guard","title":"Show HN: Phantom Guard \u2013 Detect AI-hallucinated package attacks","updated_at":"2026-03-05T23:18:44Z","url":"https://github.com/matte1782/phantom_guard"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"crop_rotation"},"comment_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["ai","hallucination","detect"],"value":"What does it even mean to detect hallucinations. The AI doesn't say something trivially false. While using GPT4 I have observed that it lies on simple things I didn't expect it to, while complex things it does very well on.
TLDR: It lies on fact based information which is mentioned in very very few places on the internet and not repeated too much. Short of having a human with the context, how do you even detect it.
Example: Ask it to describe a "Will and Grace" episode with some guest appearance. It will always make up everything including the episode number and the plot, and the plot seems very believable. If you have not watched and can't find a summary online, it is hard to say that it is a lie."},"story_title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["hallucination","detect"],"value":"SafeGPT: New tool to detect LLMs' hallucinations, biases and privacy issues"},"story_url":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"https://www.giskard.ai/safegpt"}},"_tags":["comment","author_crop_rotation","story_35652452"],"author":"crop_rotation","children":[35653546,35653829,35653881],"comment_text":"What does it even mean to detect hallucinations. The AI doesn't say something trivially false. While using GPT4 I have observed that it lies on simple things I didn't expect it to, while complex things it does very well on.
TLDR: It lies on fact based information which is mentioned in very very few places on the internet and not repeated too much. Short of having a human with the context, how do you even detect it.
Example: Ask it to describe a "Will and Grace" episode with some guest appearance. It will always make up everything including the episode number and the plot, and the plot seems very believable. If you have not watched and can't find a summary online, it is hard to say that it is a lie.","created_at":"2023-04-21T13:32:01Z","created_at_i":1682083921,"objectID":"35653469","parent_id":35652452,"story_id":35652452,"story_title":"SafeGPT: New tool to detect LLMs' hallucinations, biases and privacy issues","story_url":"https://www.giskard.ai/safegpt","updated_at":"2024-09-20T13:51:00Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"nikin_mat"},"story_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["ai","hallucination","detect"],"value":"A few weeks ago I shared an early version of MentionedBy.ai \u2014 a tool we built to monitor how your brand is mentioned inside AI-generated answers from ChatGPT, Claude, Gemini, Perplexity, and other LLMs.
We\u2019re now officially live
This started from our own frustration. Our startup wasn\u2019t being mentioned in ChatGPT responses, even for queries where we were clearly relevant. Meanwhile, legacy or SEO-heavy brands kept showing up. We realized: AI is becoming the new search engine \u2014 and no one knows how they rank in it.
So we built MentionedBy to answer questions like:
How often does your brand appear in AI-generated answers?
Is the mention accurate or hallucinated?
Are you being misrepresented, dropped, or out-ranked by competitors?
What\u2019s your sentiment score across different LLMs?
How do your rankings change over time?
Under the hood:
We run scheduled and real-time prompts via OpenRouter and native APIs
Cluster and semantically match results using OpenAI + Cohere
Log all responses in a time-series DB for trend tracking
Run hallucination detection by comparing AI claims with verified sources
We\u2019re calling this space Answer Engine Optimization (AEO) \u2014 like SEO, but for LLMs.
The platform is live:\n https://mentionedby.ai
Would love your feedback \u2014 especially on how we can make LLM visibility more transparent and actionable. AMA about the stack, prompt orchestration, hallucination detection, or what we\u2019ve learned comparing LLMs.
\u2014\nNikin\nFounder @ Synapse AI Labs\n Built in Sri Lanka, Global by default."},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Show HN: MentionedBy AI is now live"},"url":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"https://mentionedby.ai/"}},"_tags":["story","author_nikin_mat","story_44019474","show_hn"],"author":"nikin_mat","created_at":"2025-05-18T06:44:51Z","created_at_i":1747550691,"num_comments":0,"objectID":"44019474","points":1,"story_id":44019474,"story_text":"A few weeks ago I shared an early version of MentionedBy.ai \u2014 a tool we built to monitor how your brand is mentioned inside AI-generated answers from ChatGPT, Claude, Gemini, Perplexity, and other LLMs.
We\u2019re now officially live
This started from our own frustration. Our startup wasn\u2019t being mentioned in ChatGPT responses, even for queries where we were clearly relevant. Meanwhile, legacy or SEO-heavy brands kept showing up. We realized: AI is becoming the new search engine \u2014 and no one knows how they rank in it.
So we built MentionedBy to answer questions like:
How often does your brand appear in AI-generated answers?
Is the mention accurate or hallucinated?
Are you being misrepresented, dropped, or out-ranked by competitors?
What\u2019s your sentiment score across different LLMs?
How do your rankings change over time?
Under the hood:
We run scheduled and real-time prompts via OpenRouter and native APIs
Cluster and semantically match results using OpenAI + Cohere
Log all responses in a time-series DB for trend tracking
Run hallucination detection by comparing AI claims with verified sources
We\u2019re calling this space Answer Engine Optimization (AEO) \u2014 like SEO, but for LLMs.
The platform is live:\n https://mentionedby.ai
Would love your feedback \u2014 especially on how we can make LLM visibility more transparent and actionable. AMA about the stack, prompt orchestration, hallucination detection, or what we\u2019ve learned comparing LLMs.
\u2014\nNikin\nFounder @ Synapse AI Labs\n Built in Sri Lanka, Global by default.","title":"Show HN: MentionedBy AI is now live","updated_at":"2025-05-18T06:46:10Z","url":"https://mentionedby.ai/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"Shmungus"},"comment_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["ai","hallucination","detect"],"value":"Thanks for the feedback. ANGELCORE is definitely a speculative, exploratory project, so it\u2019s naturally not as grounded as my other work. I also have more traditional, technically solid projects like an LLM hallucination detector, a fragrance AI chatbot, and a machine learning model that predicts daily SEC filing numbers, all of which are finished and fully functional.
ANGELCORE is more of a grand platform where I\u2019m trying to combine a wide range of ideas, from biology to symbolic AI to physics\u2014into something new. I appreciate the call to clarity and will keep working on making the concrete aspects easier to see alongside the vision."},"story_title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["ai"],"value":"Angelcore: Building an Artificial Angel \u2013 Recursive Symbolic AI and Bio Memory"},"story_url":{"matchLevel":"none","matchedWords":[],"value":"https://github.com/Mattbusel/ANGELCORE"}},"_tags":["comment","author_Shmungus","story_44195227"],"author":"Shmungus","comment_text":"Thanks for the feedback. ANGELCORE is definitely a speculative, exploratory project, so it\u2019s naturally not as grounded as my other work. I also have more traditional, technically solid projects like an LLM hallucination detector, a fragrance AI chatbot, and a machine learning model that predicts daily SEC filing numbers, all of which are finished and fully functional.
ANGELCORE is more of a grand platform where I\u2019m trying to combine a wide range of ideas, from biology to symbolic AI to physics\u2014into something new. I appreciate the call to clarity and will keep working on making the concrete aspects easier to see alongside the vision.","created_at":"2025-06-05T20:25:55Z","created_at_i":1749155155,"objectID":"44195462","parent_id":44195252,"story_id":44195227,"story_title":"Angelcore: Building an Artificial Angel \u2013 Recursive Symbolic AI and Bio Memory","story_url":"https://github.com/Mattbusel/ANGELCORE","updated_at":"2025-06-05T20:27:54Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"cobusgreyling"},"comment_text":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["ai","hallucination","detect"],"value":"Balancing Latency, Interpretability, and Consistency in Hallucination Detection for Conversational AI
If you find any of my observations to be inaccurate, please feel free to let me know\u2026
I appreciate that this study focuses on introducing guardrails & checks for conversational UIs.
When interacting with real users, incorporating a human-in-the-loop approach helps with data annotation and continuous improvement by reviewing conversations.
It also adds an element of discovery, observation and interpretation, providing insights into the effectiveness of hallucination detection.
The architecture presented in this study offers a glimpse into the future, showcasing a more orchestrated approach where multiple models work together.
The study also addresses current challenges like cost, latency, and the need to critically evaluate any additional overhead.
Using small language models is advantageous as it allows for the use of open-source models, which reduces costs, offers hosting flexibility, and provides other benefits.
Additionally, this architecture can be applied asynchronously, where the framework reviews conversations after they occur. These human-supervised reviews can then be used to fine-tune the SLM or perform system updates."},"story_title":{"matchLevel":"none","matchedWords":[],"value":"[dead]"}},"_tags":["comment","author_cobusgreyling","story_41444851"],"author":"cobusgreyling","comment_text":"Balancing Latency, Interpretability, and Consistency in Hallucination Detection for Conversational AI
If you find any of my observations to be inaccurate, please feel free to let me know\u2026
I appreciate that this study focuses on introducing guardrails & checks for conversational UIs.
When interacting with real users, incorporating a human-in-the-loop approach helps with data annotation and continuous improvement by reviewing conversations.
It also adds an element of discovery, observation and interpretation, providing insights into the effectiveness of hallucination detection.
The architecture presented in this study offers a glimpse into the future, showcasing a more orchestrated approach where multiple models work together.
The study also addresses current challenges like cost, latency, and the need to critically evaluate any additional overhead.
Using small language models is advantageous as it allows for the use of open-source models, which reduces costs, offers hosting flexibility, and provides other benefits.
Additionally, this architecture can be applied asynchronously, where the framework reviews conversations after they occur. These human-supervised reviews can then be used to fine-tune the SLM or perform system updates.","created_at":"2024-09-04T12:28:00Z","created_at_i":1725452880,"objectID":"41444852","parent_id":41444851,"story_id":41444851,"story_title":"[dead]","updated_at":"2024-09-20T17:45:04Z"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"matt_d"},"title":{"fullyHighlighted":false,"matchLevel":"partial","matchedWords":["hallucination","detect"],"value":"LettuceDetect v2 in Semantic Router: Gen. Hallucination Detection vLLM Endpoint"},"url":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["ai","hallucination","detect"],"value":"https://vllm-sr.ai/blog/lettucedetect-v2-generative-hallucination-detection/"}},"_tags":["story","author_matt_d","story_49201431"],"author":"matt_d","created_at":"2026-08-06T19:45:23Z","created_at_i":1786045523,"num_comments":0,"objectID":"49201431","points":1,"story_id":49201431,"title":"LettuceDetect v2 in Semantic Router: Gen. Hallucination Detection vLLM Endpoint","updated_at":"2026-08-06T19:49:56Z","url":"https://vllm-sr.ai/blog/lettucedetect-v2-generative-hallucination-detection/"},{"_highlightResult":{"author":{"matchLevel":"none","matchedWords":[],"value":"priyankanath"},"title":{"fullyHighlighted":false,"matchLevel":"full","matchedWords":["ai","hallucination","detect"],"value":"Why is this misleading?: Detecting News Headline Hallucinations by AI"},"url":{"matchLevel":"none","matchedWords":[],"value":"https://vevesta.substack.com/p/everything-you-need-to-know-about-660"}},"_tags":["story","author_priyankanath","story_39824613"],"author":"priyankanath","created_at":"2024-03-26T05:52:36Z","created_at_i":1711432356,"num_comments":0,"objectID":"39824613","points":1,"story_id":39824613,"title":"Why is this misleading?: Detecting News Headline Hallucinations by AI","updated_at":"2024-09-20T16:39:14Z","url":"https://vevesta.substack.com/p/everything-you-need-to-know-about-660"}],"hitsPerPage":20,"nbHits":234,"nbPages":12,"page":0,"params":"query=AI+hallucination+detect&advancedSyntax=true&analyticsTags=backend","processingTimeMS":17,"processingTimingsMS":{"_request":{"roundTrip":17},"afterFetch":{"format":{"total":1},"merge":{"mergeLoop":{"prepareNextHit":1,"total":1},"total":2},"total":2},"fetch":{"query":7,"scanning":6,"total":14},"total":17},"query":"AI hallucination detect","serverTimeMS":19}