Question

Passage A

The profound disruptive potential of autonomous algorithmic trading systems poses a formidable challenge to traditional legal paradigms, particularly those of criminal jurisprudence. At the core of criminal liability lies the principle of *mens rea*, or a "guilty mind." A conviction typically requires demonstrating that a defendant possessed the requisite intent to commit a crime. Autonomous algorithms, however, operate absent consciousness or volition; they merely execute complex mathematical optimizations. Consequently, when an algorithmic trading system independently orchestrates a catastrophic "flash crash," the traditional search for *mens rea* hits a dead end. Attempting to attach criminal liability to the human programmer is equally problematic, as it would require proving the programmer specifically coded the algorithm with the intent to manipulate the market—a rare occurrence given that most such disruptions arise from unforeseen interactions in live trading environments. Therefore, shoehorning algorithms into traditional criminal statutes inevitably fails. Furthermore, attempting to personify software to satisfy criminal statutes risks creating dangerous legal precedents that could be misused in other domains of technology law. The most logically sound remedy is to eschew criminal law entirely in these instances and instead rely on corporate strict liability within civil law. By holding the financial institution unconditionally responsible for the financial damages caused by its proprietary software, the legal system ensures that victims are compensated without warping the foundational tenets of criminal jurisprudence to accommodate non-human actors.

Passage B

As autonomous systems increasingly mediate financial markets, the traditional liability binary—requiring either demonstrable human intent or defaulting to strict liability—proves inadequate. Proponents of strict liability argue it provides a clean solution for algorithmic harms by ensuring compensation without the messy business of proving intent. However, this approach is fundamentally flawed when applied to complex artificial intelligence. Strict liability effectively treats the devastating impacts of algorithmic emergent behaviors merely as an operational cost of doing business, completely bypassing the crucial concept of fault. We need not pretend that an algorithm possesses a human mind to recognize that the reckless deployment of opaque, inadequately tested systems constitutes a breach of duty. The law must evolve to recognize a standard of "algorithmic negligence." Under this framework, liability would attach to the human operators and institutions that deploy these systems, not because they intended the specific harm, but because they failed to adequately monitor, test, or constrain foreseeable emergent risks. By focusing on the deployment process rather than the algorithm's internal logic, this fault-based approach provides a powerful deterrent against the reckless use of autonomous trading systems, penalizing negligence without demanding anachronistic proofs of *mens rea*. By treating the deployment of advanced AI as a specialized domain requiring rigorous professional standards, the legal system can adapt to technological realities without sacrificing the core moral functions of the law.

Based on the passages, which of the following statements accurately characterize the authors' respective positions on algorithmic trading and legal liability? (Select all that apply)