ElevenLabs

Discussion around ElevenLabs over the past four weeks has been dominated by frustration rather than enthusiasm, with bugs and reliability emerging as the heaviest themes in the conversation by a significant margin. Bug-related mentions outpaced even the next largest complaint category by nearly two to one, and reliability concerns added further weight to a picture of users struggling with day-to-day dependability. Pricing drew recurring criticism as well, and a thread of privacy concern ran through the discussion, particularly energized by announcements around AI voice partnerships using the likenesses of public figures. At least one commenter reacted to the Stan Lee partnership with pointed hostility, describing the practice in morally charged terms, suggesting that ethics around voice and likeness use touched a nerve for a subset of the audience.

On the positive side, commenters credited the product on feature quality and AI output quality, and there was genuine enthusiasm in some corners around integrations and competitive positioning. Feature requests appeared on both the praise and complaint sides, indicating an engaged user base that sees potential but wants more. A competitor comparison thread suggested users are actively benchmarking the product, which cuts both ways in terms of sentiment.

The score trajectory tells a clear story of a sharp decline from a relatively strong opening. Sentiment was meaningfully higher in late April and early May, then dropped hard around the week of May 11 as mention volume surged, suggesting a specific event or release drew a wave of critical attention. Scores have largely plateaued in the low-to-mid forties since then, with only modest week-to-week variation, indicating that initial negativity has stabilized rather than recovered.

Opinion was most divided around the ethics of AI voice cloning and likeness use, where reactions ranged from celebratory to strongly condemnatory. The investment news from Poland generated more neutral or positive framing, while feature request threads reflected a split between users who find the current toolset capable and those who feel it falls short of practical workflow needs.