This work examines the relationship between listed companies and high frequency trading (HFT). First, price-formation process performed by trading venues is analysed, taking into account micro structural changes produced by high frequency trading. In this respect, it is demonstrated that equity trading referable to HFT activity cannot be based on information provided by companies in order to be compliant with ad hoc and on-going mandatory disclosure requirements. It follows that probably the most important rationale for imposing mandatory disclosure rules is severely undermined, because at present a consistent percentage of transactions in shares (the large majority in the US) does not and cannot take into account the costly amount of information disclosed by listed companies. Thus, the role of mandatory disclosure requirements is shrinking. Second, HFT reveals and produces structural changes in modern equity markets, undermining also those rationales for prohibiting market abuse which are based on theories linked to market efficiency and to the price-formation process. On the other hand, egalitarianism and fiduciary relationship acquire renewed centrality as alternative rationales to impose sanctions in case of insider dealing. As a matter of fact, high frequency traders base their trading decisions almost exclusively on price sensitive data which are bound to remain unusable (sometimes even undisclosed) to the rest of the market. Third, high frequency traders, differently from algorithmic traders, embody a new type of fully automated trading decision maker in financial markets, whose ‘behaviour’ is structurally different with respect to those of classic market participants. This circumstance sheds new light on the on-going debate regarding rationality in financial markets, which in the recent past was dominated by the contrast between Behavioural Finance and Efficient Capital Market Hypothesis (ECMH). Fourth, HFT lessens classic (low-frequency) market participants' incentives to discover and analysed information, since it generates self-similarity and reflexivity in automated markets, increasing volatility and instability. It is then rational, for a long term oriented investor, to discount securities by the risk associated with market crashes. The latter are represented not only by the Flash Crash happened the 6th of May 2010, but also by an increasing number of computer-related crashes referable to single stocks and by thousands of mini flash-crashes. Fifth, the latest regulatory measures dealing with HFT are analysed in order to assess their efficacy with respect to the issues mentioned above.