随着人工智能技术的蓬勃发展，许多公司它们能开发出无所不能的AI。
With the vigorous development of artificial intelligence technology, many companies claim they can develop AI capable of doing anything.
但一个核心问题也随之浮现：我们应该如何确保AI的行为符合人类的道德标准？
But a core question also emerges along with it: How do we ensure that AI's behavior conforms to human moral standards?
虽说技术进步能带来巨大效益，但如果不加以约束，其后果可能十分。
Although technological progress can bring enormous benefits, if left unchecked, its consequences could be extremely cruel.
例如，一个用于法庭的AI系统，如果它的数据库里充满了偏见的数据，它就可能做出且不公平的判决。
For example, an AI system used in court, if its database is full of biased data, may hand down arbitrary and unfair judgments.
这不仅仅是一个技术问题，更是一个涉及社会阶级和公平的深刻伦理问题。
This is not just a technical problem, but more deeply an ethical issue involving social class and fairness.
因此，工程师们必须，在设计AI之初就把伦理考量精心地融入其中。
Therefore, engineers must put in the effort to carefully incorporate ethical considerations from the very start of designing AI.
我们需要制定明确的规则，比如通过民主来决定AI的某些行为边界。
We need to establish clear rules, for example using democratic voting to decide certain behavioral boundaries of AI.
这些规则应在AI的源代码中，并且能在实地测试中得到验证。
These rules should be attached within the AI's source code, and be verifiable in field testing.
甚至，在一些高度敏感的领域，如军事或护理，我们可能需要更严格的审核与。
Even more, in some highly sensitive fields, such as the military or healthcare, we may need stricter review and approval.
同时，公众对AI的教育也至关重要。
At the same time, educating the public about AI is also crucial.
我们不能让AI的毫无约束地燃烧，以免它烧毁我们赖以生存的社会信任。
We cannot let the flame of AI burn without restraint, lest it burn away the social trust that we rely on to live.
通过共同努力，我们或许能开发出既聪明又富有同情心的AI伙伴。
Through joint effort, we may perhaps be able to develop AI companions that are both intelligent and full of compassion.
这样的AI将真正服务于人类，而不是成为威胁。
Such AI will truly serve humankind, rather than becoming a threat.
这或许是我们在AI时代最重要的使命。
This is perhaps our most important mission in the age of AI.