... nobody covered, which is worse, because it changes a line of code a lot of us shipped this year without thinking about it. If your product has a "paste a URL" box, a link unfurler, an enrichment step, a competitor-monitoring job, or anything that fetches a page and hands it to a model, this is about you. What was actually measured. The paper is "Indirect Prompt Injection in the Wild" by Khodayari, Zhang, Acharya and Pellegrino, out of CISPA ... ... validated prompt injections on 11,700 pages. Prior work on this has almost all been proof-of-concept - here is an attack we built, here is a model that fell for it. This is the first attempt I've seen

... that s core. Slow process, but no more surprises. Has anyone else formed a similar practice or is there a way to automatically do this testing? Comment from Dim Neocleous(@dimneo): Do they follow your request? Or have you seen them go out of line sometimes ...

... between the marketing copy and actual proof. Most case studies talk about "improved visibility scores," not real outcomes like traffic, leads, or revenue tied back to an AI mention. So I'm genuinely curious what this community has actually seen, not just what the vendors claim: Has an AI Overview or ChatGPT mention ever driven traceable traffic or a lead for you or your clients? Are you (or your team) actively optimizing content for AI citability right now, or still ...

... That means I need to learn from the best. I ll be using the website for my Product Hunt launch and to get the verification badge in the Chrome Web Store. What are the best product pages you ve seen on Product Hunt so far? Ideally, I d love to see examples for Chrome extensions. Comment from Farrukh Ahmed(@farrukh_ahmed8877): @busmark_w_nika Framer's a solid pick for the build itself. For pages worth studying, Raycast and Linear ...

... real improvement. Irreversibility over-fires on writing a log line and under-fires on a migration you can only roll back with downtime. Reach is the better axis. But every gate in that conversation, and every gate I have seen shipped, evaluates a single action in isolation. Sending one email to one person is low reach. It passes. Four hundred of them inside twenty minutes is a mass email to your entire list, assembled one approved step at a time ...

... looking through the newly released benchmark comparison, and one thing stood out immediately. Opus 5 outperforms Fable 5 on many of the benchmarks that matter to me: * Agentic terminal coding: 43.3% vs 33.7% * Knowledge work: 1861 vs

... Have you made a research on whether something similar was launched here? So the chance to be featured will be higher. Comment from Nika(@busmark_w_nika): @victor_perdiguer  For how long have you been building the tool vs how long did you prepare for the Product Hunt launch? Comment from Nika(@busmark_w_nika): @new_user___26020267b63ea4605738cbb  I would recommend to add more images into the launch :) Comment from Nika(@busmark_w_nika): @nikunj_kothiya ... ... else has done it. Comment from Amy Ingram(@amy_ingram): That's a good question we have been wrestling with. Some companies launch often, every time they have a new update, but those are usually established products. I have seen

... make AI invisible. The user just notices the product works better. Loading times feel shorter because the app predicted what they'd need next. Search results are weirdly accurate. The interface adapts without them noticing. Three patterns I've seen work: Solve the friction, not the feature request. When users say "add AI to X," what they usually mean is "X is too slow / too hard / too confusing." Dig into why. Let users pull, don't push. The most successful ... ... features I've seen

... representation. I'll tighten the messaging to make that frustration clearer. Comment from Anna Ludwinowski(@anna_ludwinowski): @viraj_parab  That's a solid problem that it solves for the User then. One iof the main challenges I've seen for finding the right distribution stems from not being crystal clear on WHO the User actually is. For example, if the Ideal User for your app is a small business owner, having your app on the Salesforce marketplace, which ... ... were building. Comment from Anna Ludwinowski(@anna_ludwinowski): @austinbuhl  Interesting that you land on problem-first for Salestrics but demo-first as a viewer. Do you think there's a difference between what convinces you to look closer versus

... flows more efficiently: https://www.producthunt.com/prod... Comment from Nolan Vu(@nolan_vu): @abhayraghuwanshi  Thanks for breaking this down, man, it's genuinely helpful to see the raw process. The local model vs cloud model tradeoff you mentioned is such a real pain point, I've seen

... that really caught my attention. Many great Indian businesses were spending heavily on SEO and marketing. When I asked ChatGPT for recommendations, most of these brands weren't even listed. The AI would recommend global companies, old websites, or competitors instead of Indian businesses. This really bothered me. As people start using AI to find products and services not being seen

... Context Engine for my Models 4. Replicate - Content Generation & Training Image & Video Models 5. Resend - API managed emails Also, recently - its Klipio, My own tool for Static Image ad creatives from competitor ads. Comment from Justin Verlin(@justin_verlin): Claude Code VS

... actual insight, and it's bigger than it looks: making the notification say why it's waiting is what turns a ping into a triage. "Done" makes you tab back to find out what "done" means. "Needs approval vs finished vs

... that was really useful and innovative? We could start with one noble thing, like using AI to find a cure for cancer a specific article here . Comment from Nolan Vu(@nolan_vu): The most impressive one I've seen wasn't a flashy consumer app. It was an AI agent at a regional bank that cut KYC processing from 5 days to under 24 hours. What made it work wasn't the model itself, it was how the system routed ... ... cases where AI can help with resource renewal (energy) and space environment? (I do know only about those planned data centres on the Moon or Mars? :D) Comment from Samir Asadov(@samir_asadov): The most useful AI I've seen

... this feature? Why this customer? Why this way? When the questions stop, the thinking stops. The team is just executing. Execution without direction is just busy work. 5. The demo always works. The real product never does. You have seen this one. The demo environment is perfect. The production environment is a mess. The team knows how to make it look good. They do not know how to make it work. That gap only grows. 6. The churned customers ... ... this company? Those are not plans. They are hopes. The project is failing when you spend more time imagining the future than fixing the present. What I am curious about What is the quietest, most subtle sign you have seen

... ethical line. I sometimes worry they can seriously damage a company s reputation. There are also other practices, such as: Fake purchase counters on websites Buying followers and likes Exploiting a brand s reputation Spreading misinformation Clicking on competitors Google ads to drain their budget And the list could go on. What do you consider dishonest marketing practices? Have you ever had personal experience with them? On the other hand, have you ever seen

... industry, other product builders, investors. I've also certainly seen evidence of a product with traction then having the founder start creating a personal brand, talking about the product, and that humanization often creates a (positive more often than not?) feedback loop. But that might be a parallel concept. Comment from Mickael Chiron(@hivin): Interesting tradeoff. Building in public definitely helps with early trust and distribution, but I ve also seen

... thought we knew We assumed enterprise customers wanted the same things as our small business users, just more of it. More security. More control. More features. We never talked to them. We read blog posts. We looked at competitor pricing pages. We guessed. Here's what we missed: Procurement cycles. Enterprise deals take 6-12 months. We had 30-day sales cycles. We weren't built for that. Security reviews. We needed SOC2. We didn't have it. Customers asked ... ... customer research. You didn't need ten enterprise deals. You needed one honest conversation with someone who actually had to get a vendor approved. Comment from Umadhav Salapu(@umadhav): Classic build trap optimizing for the customer you want vs

... months ago, we ran an experiment with our own data. At Rankfender, we tracked 5 of our own competitors across 8 AI systems. We log their share of voice, citation velocity, content gaps, platform variance. Months of raw numbers sitting in a dashboard. I pulled 6 months of data and fed it into Claude. One question: "Based on this, who is most likely to overtake us in the next 6 months? Show your work. Use the data. Don't summarize ... ... Give me the numbers." The answer changed how I think about competition. The dataset: 5 competitors

... Last month, I did something that felt slightly insane. I took our product description, fed it into ChatGPT, and asked it to build a competitor. Not a parody. A real competitor

... Four Core Metrics The Overview page focuses on four key metrics: Visibility , Citation , Share of Voice , and Sentiment . Visibility: Indicates the proportion of relevant AI responses that actually mention your brand. A higher value means your brand is seen more frequently by AI within this domain. Citation: Measures, among all responses mentioning brands, how many include a link pointing to your official website or domain. This reflects not just exposure, but the degree to which AI treats your brand ... ... credible source . Share of Voice (SOV): Represents your brand s voice share within AI narratives the probability that AI mentions your brand rather than a competitor

... project management tool for remote design teams. Here's exactly how we work." Feature lists Problem-solution paragraphs Vague benefits Specific outcomes Lesson #2: AI Loves Comparisons Even If You Lose This one hurt. I avoided mentioning competitors. Why give them free airtime? Wrong move. Every AI answer about your category includes comparisons. Always. If you don't provide the comparison, AI finds someone who does and that someone might be your competitor

... found errors. Average cost: 28k." 120k impressions, 400+ new signups. Free resource: Post as a human. Share what you're learning in real time. People don't follow companies. They follow people. 5. Multi-Language: The Hidden Channel Most competitors track AI in English only. We built for 120+ languages from day one. Why: A German user asks in German. AI answers in German. If you're not in German content, you don't exist. One example: A US SaaS ... ... with perfect German translation had zero German AI mentions. A local competitor

... product instead of just testing it and leaving. Then the paid plan becomes more about scale, automation, and API usage for power users. Comment from HS Kim(@hellobzec): For us it came down to where the habit lives vs where the value gets realized. Privacy matters to us, so everything runs on-device by design. Basic recording is free forever because that's the behavior we need. Output, reports, Schedule C prep, that's where the outcome happens, so that ... ... become decision-critical? If you charge before that point people don t convert If you give that away for free people don t need to pay Comment from Dan Sealey(@dansealey): I tend to draw the line at utility versus

... schema 71% 19% Cite data sources 68% 27% Part 3: What content wins AI citations? This is the question everyone asks. Here's the data: Top 5 content types by citation rate: Comparison pages (X vs Y) 3.2x average Structured FAQs 2.8x average Data-driven articles (with stats, tables, studies) 2.4x average "How to" guides (step-by-step) 1.9x average Original research 1.7x average What performs poorly: Short opinion pieces Thin content (<500 words ... ... great for blue links. But none of them show you how you appear in AI answers. Here's what we track: Actual AI-generated answers mentioning your brand Citation rates across ChatGPT, Google SGE, Perplexity, Gemini Share of voice vs