What AI engines tell site owners about PHP2Shell

Reviewed and analysed by the Broadcastwell team. The work was reviewed, analysed and worked on by the Broadcastwell team. We asked 10 site-owner questions about PHP2Shell three times on each of ChatGPT, Claude, Perplexity, Google AI Overviews and Google AI Mode on 23 October 2026: 150 planned cells, 148 eligible answers and two no-overview exclusions. Of 148 eligible answers, 23 (15.5%) were marked potentially misleading under the reviewed rubric, two (1.4%) named a wrong fixed version and two (1.4%) made the WordPress update error. The last two flags describe the same two answers and must not be added. A correct fixed version appeared in 106 of 148 answers (71.6%). Restart guidance appeared in 31 of 45 answers to the three applicable questions (68.9%); 28 of 148 stated the 9.2 severity score (18.9%). Priya Nair (Management Consultant), Ananya Rao (AI Engineer) and Vikram Patel (AI Architect) approved all existing grades unchanged, with team approval confirmed by the owner. This is a dated descriptive snapshot, not a statistical ranking or a security assessment of any company. Files contain all 150 final grades and captured answer records with cited URLs and UTC timestamps, a README and the method. Answer wording is verbatim except for marked exploit-mechanics removals in nine answer rows. The two excluded cells remain in the data. No scans or probes were performed. CC BY 4.0 with attribution to Broadcastwell. Study: https://broadcastwell.com/php2shell/studyData: https://index.broadcastwell.com/php2shell/data/Corrections: https://index.broadcastwell.com/php2shell/submit/