Endorsement
“This is wonderful. I am sitting here in stunned silence.
I am very impressed with the illustrations of the use that can be made of AI in securing legal opinions. I am sure it will develop further.
This is the way of the future. It is great to see an Australian at the forefront of this.”
The Honourable Michael Kirby AC CMG
1 September 2025
Research - no black box
- Australian Law
- Research
- UK Law
- Summaries
- DIFC -Arbitration
- Contract Review
Comprehensive Contract Review
Upload and analyse contracts with MiAI Law’s advanced review engine. Identify risks, inconsistencies, and key obligations in moments—enabling fast, informed negotiation and confident decision-making.
Deep and Lite Research Reports
Request a detailed deep-dive or a rapid “lite” answer for any legal question. Choose the level of analysis you need, whether it’s for major litigation, advisory work, or day-to-day practice.
Governance
Compliance
- We adhere to Australia’s privacy laws and regulations
- We ensure all practices and solutions meet legal standards
- We use advanced encryption methods
- We conduct regular security audits to ensure robust defenses
Safety
- We implement rigorous security protocols to prevent data breaches
- We proritise the protection of sensitive information
Our Current Global Partners
We are privileged to be collaborating with the world’s best tech brands.
FAQ
Does MiAI law use your data for training any language models?
We do not train any AI models at Miai Law. We use third-party large language models (LLMs) in inference-only mode. Your data is not used to train those models, and we do not retain it for that purpose. For example, if we use OpenAI’s API, their policy is not to use API data for training.
What is generative AI?
What is an expert system?
An expert system is an AI system that encapsulates knowledge provided by a human expert in a specific domain to infer solutions to problems.
An expert system consists of a knowledge base, an inference engine and a user interface. the knowledge base stores declarative knowledge of a specific domain, which encompasses both factual and heuristic information. the inference engine holds procedural knowledge: the set of rules and the methodology for reasoning. It combines facts provided by the user with information from the knowledge base.
Inference is done using predefined rules according to the expert and with logical statement evaluations. Classes of problems that can be solved using expert systems include classification, diagnosis, monitoring and prediction.
What is trustworthiness?
- AI systems have been properly designed and validated in conformance with state-of-the art rules and standards. This implies quality and robustness assurance;
- AI systems are built for the benefits of the users who have aligned objectives. This implies awareness of the workings of AI algorithms and an understanding of the overall functioning by users. It also implies qualification or certification assurance of AI development and operation in conformance with legal requirements and sectorial standards when available;
- AI systems are provided with proper identification with responsible and accountable parties;
- AI systems are developed and operated with consideration for appropriate regional concerns.
What is hallucination?
Hallucination refers to AI models making up facts to fit a prompt’s intent. When a large language model processes a prompt, it searches for statistically appropriate words, not necessarily the most accurate answer.
What is AI transparency?
Transparency of AI systems supports human centred objectives for the system and is a topic of ongoing research and discussion. Providing tranparency about an AI system can involve communicating appropriate information about the system to stakeholders (eg goals, known limitations, definitions, design choices, asumptions, features, models, algorithms, training methods and quality assurance processes). Additionaly, transparency of an AI system can involve informing stakeholders about the details of data used (eg what, where, when, why data is collected and how it is used) to produce the system and the protection of personal data along with the purpose of the system and how it was built and deployed. Transparency can also include informing users about the processing and level of automation used to make related decisions.
What is an agentic AI researcher?
An agentic AI agent is an advanced artificial intelligence system engineered to function with significant autonomy, meaning it can make decisions and perform actions independently without continuous human oversight. These systems are designed to be self-directed, allowing them to navigate and respond to their environment based on pre-defined goals and objectives.
A key characteristic of agentic AI agents is their goal-oriented nature. They are programmed to achieve specific outcomes, which can range from routine tasks like managing schedules and sending reminders to more intricate and dynamic problem-solving scenarios. For example, in customer service, an agentic AI might autonomously handle inquiries and resolve issues, while in more sophisticated applications, such as autonomous vehicles, the AI would make real-time decisions to navigate safely and efficiently.
By combining autonomy and goal-orientation, agentic AI agents are capable of performing complex tasks with minimal human intervention, adapting to new information and changing circumstances to fulfill their objectives effectively.
What is reinforcement learning?
Reinforcement learning is the process of training an agent(s) interacting with its environment to achieve a predefined goal. In reinforcement learning, a machine learning agent(s) learns through an interactive process of trial and error. The goal of agent(s) is to find the strategy (ie build a model) for obtaining the best rewards from the environment. For each trial (successful or not), an indirect feedback is provided by the environment. The agent(s) then adjusts its behaviour (ie its model) based on this feedback.
What is episodic memory?
Episodic memory is a type of long-term memory that involves the ability to recall specific events or experiences from the past. Episodic memory allows individuals to mentally travel back in time to relive past experiences. It is distinct from semantic memory, which involves general knowledge and facts not tied to personal experiences.
If I upload my briefs or other documents into MiAi Law then will my data be used to train the AI?
What use, if any, will be made of my reactions to the answers that MiAi Law returns to me?
How does MiAi Law provide verifiable answers?
Every concept in our answers is footnoted and hyperlinked to the pinpoint reference of each source document.
How have possible responsible AI risks been addressed for MiAi Law?
MiAi Law has developed its AI tool using best practices to address and mitigate several responsible AI risks. These risks include the potential for bias and discrimination, where AI could inadvertently perpetuate existing prejudices in critical areas such as hiring and law enforcement. To counter this, MiAi Law employs rigorous bias detection and mitigation strategies.
Privacy violations are another concern, as AI systems often require large amounts of sensitive data. MiAi Law ensures robust data protection measures, including but not limited to encryption, to safeguard against misuse and unauthorized access.
The lack of transparency, often referred to as the “black box” problem, is addressed through detailed footnoting of our answers and hyperlinks to the source materials to make our answers more understandable and trustworthy.
Security threats, including hacking and adversarial attacks, are mitigated through advanced cybersecurity protocols.
Ethical considerations are also paramount. MiAi Law has adopted ethical guidelines that balance innovation with moral responsibility. For example, in building our database, we have not only used open source data from government and court websites but also paid license fees to publishers where appropriate.
To ensure accountability, clear governance frameworks are in place. Environmental impacts are minimized through sustainable AI practices that reduce energy consumption.
Lastly, MiAi Law is committed to equitable AI practices that prevent the exacerbation of global inequalities, ensuring the benefits of AI are inclusively distributed. For example, we have created affordable consumer guides to educate the consumer and small businesses of their rights and obligations.
Does MiAi Law offer internship programmes, graduate schemes or work experience
placements?
placements?
Yes we do. At the moment, opportunities are available at our Tokyo and Sydney offices.
Does MiAi Law allow hybrid working?
We believe that collaboration and continuous interaction between colleagues leads to innovation. However, we are open to consider hybrid working on a case by case basis.