Sorbonne has for the first time in the university's history made a complaint against two students after having cheated on an exam. The two students have supposedly got help from a third person to correct their papers.

Two students in their first year of law studies have paid 50 euros each in bribes to have their exam paper corrected. They have most likely used their cellular phone to contact a third person who has then corrected their papers.

Several hundreds of students cheat on exams every year and the methods and imaginations are better than ever; programmed calculators; copying from Internet; fake identity card and so on. However, the sanctions can be up to two years exclusion from sitting an exam.

The National Council for Higher Education and Research (Cneser), which is an administrative jurisdiction, treats every year cases of fraud in the French education system.

809 procedures against university students have been raised because of cheating. In 1999, 84 cases of fraud were detected for the graduating exam, against 148 cases in 2003. Out of the 77 cases discovered in 2004, 55 concerned use of personnel documents; 8 use of mobile phones; 6 change of draft and 8 false identity.

According to the report of the general inspection of the administration of the national education and the research (IGAENR), the evaluation of the students at the university is not good. Actually, the fragmentation of the evaluations (due to the transition to the half yearly of the studies connected to the passage in the LMD) and the complexity of rules, return the illegible system for the students. It also entails disparities of treatment; thus universities develop their own rules of evaluation: the faculties with big workforce opt for the multiple choice question paper, faster and easy to organize. Besides, the cheating is another factor that undermine the credibility of the diplomas: according to the questioned students, between 25 and 50 % of the students resort to it.

The ministry of Higher education and research entrusted to the national Centre of scientific research (CNRS) a mission on scientific integrity. Scientific fraud is varied: biased manufacturing or forgery of results, biased interpretation or selection of data, alteration of curves or images, plagiarism, theft of ideas, financial profit-sharing's ... The increasing pressure which practices on the researchers (among which the career and the credits depend strictly on the quantity of articles and on produced results) tend to multiply the fraudulent practices.

An investigation published by Nature, reveals that a scientific plagiarism has course among researchers too. Two researchers from South-western Medical Centre of the University of Texas, Mounir Errami and Harold Garner, auscultated an American medical documentation base, Medline, where summaries of 17 million articles published in more than 5,000 reviews from some 80 countries are indexed. They screened it through an engine search, eTBLAST, which is able to locate the "similarities". While focusing on the most quoted 7 million articles, they located a little more than 70,000 cases of "high resemblance". That which, taking into account the limits of the software, makes them estimate the number of plagiarisms at more than 200,000, out of the 17 million referred articles.