LLM Targeted Underperformance Disproportionately Impacts Vulnerable Users · HackerPapers discussions

LLM Targeted Underperformance Disproportionately Impacts Vulnerable Users — mentioned by yogthos

Wed, 15 Jul 2026 14:05:41 GMT

While state-of-the-art large language models (LLMs) have shown impressive performance on many tasks, there has been extensive research on undesirable model behavior such as hallucinations and bias. In this work, we investigate how the quality of LLM responses changes in terms of information accuracy, truthfulness, and refusals depending on three user traits: English proficiency, education level, and country of origin. We present extensive experimentation on three state-of-the-art LLMs and two different datasets targeting truthfulness and factuality. Our findings suggest that undesirable…

Paper and discussion history. A mention is not an endorsement.

LLM Targeted Underperformance Disproportionately Impacts Vulnerable Users — mentioned by yogthos

Sat, 13 Jun 2026 10:32:21 GMT

I'll take these 'authoritarian' models from China any day over whatever you call this https://arxiv.org/abs/2406.17737

Paper and discussion history. A mention is not an endorsement.

LLM Targeted Underperformance Disproportionately Impacts Vulnerable Users — mentioned by yogthos

Sun, 24 May 2026 13:59:50 GMT

Reads to me like Anthropic just wants to ban competition from open models coming out of China which are eating their lunch. And the actual race isn't with compute availability, it's with energy production. Energy costs in China are already 3x lower than in the US, and that makes running models and scaling them out a lot more affordable. Meanwhile, the US grid already has only something like 30% leftover capacity, and if that gets gobbled up then there's no slack available for events like heat waves or other natural disasters. So, there's literally no room to grow. And building out new…

Paper and discussion history. A mention is not an endorsement.

LLM Targeted Underperformance Disproportionately Impacts Vulnerable Users — HN submission

Thu, 14 May 2026 14:44:19 GMT

LLM Targeted Underperformance Disproportionately Impacts Vulnerable Users

9 points; 3 discussion replies on HN. Paper and source history. A submission is not an endorsement.

LLM Targeted Underperformance Disproportionately Impacts Vulnerable Users — HN submission

Thu, 16 Apr 2026 14:32:14 GMT

LLM risk spreading misinformation to humans who are least able to identify it

4 points; 1 discussion reply on HN. Paper and source history. A submission is not an endorsement.