Machines execute 85% of Australian share trading. ASIC is tightening the rules

Australia's financial markets have undergone a remarkable transformation. Computers now execute an estimated 85% of trading in Australian listed equities. In some futures markets, the proportion is even higher. ASIC has responded by introducing new rules to manage the risks of increasingly automated markets.

What is algorithmic trading?

Algorithmic trading uses computer programs to decide when, how and sometimes whether to buy or sell financial instruments.

An investment fund purchasing $100 million of shares, for example, might use an algorithm to spread its purchases across thousands of transactions, reducing their impact on market prices. Other algorithms exploit small price differences between markets or continuously provide buying and selling quotes.

High-frequency trading takes automation further, executing transactions in milliseconds or faster.

Importantly, algorithmic trading is not synonymous with artificial intelligence. Conventional algorithms follow programmed rules, whereas AI-enabled systems can identify patterns and adapt their behaviour using machine learning.

Which markets are affected?

ASIC estimates that algorithms account for 85% of Australian listed-equity trading, 94% of SPI 200 futures trading and 46% of three-year Treasury bond futures trading.

These markets are central to Australia's financial system. Equities facilitate investment in companies, share-index futures allow investors to hedge market exposure, and bond futures help financial institutions manage interest-rate risk.

The figures measure algorithmic trading activity, not the proportion of investors using AI.

What has ASIC found?

Automation can improve liquidity, reduce transaction costs and facilitate faster execution. However, it introduces risks that traditional market supervision was not designed to address.

Faulty algorithms can generate large volumes of erroneous orders. Interactions between automated systems may amplify price movements during volatile trading. AI models introduce additional uncertainty because their decisions can be difficult to explain or predict.

ASIC has identified these as potential threats to orderly markets, rather than evidence that automated trading is inherently harmful.

What is ASIC doing?

On 24 September 2026, ASIC finalised amendments to its Market Integrity Rules covering securities and futures markets.

The changes strengthen requirements for market participants to develop, test, monitor and govern trading systems and algorithms. They also clarify the prohibition on trading that creates a false or misleading appearance in financial markets, including AI-enabled activity.

The technology-neutral rules commence on 18 March 2028, following an 18-month transition.

ASIC is simultaneously consulting on simpler regulatory guidance, proposing to reduce its volume for securities-market participants by almost 60%. This reduction is proposed, not yet implemented.

The underlying regulatory question is increasingly important: as computers undertake more trading decisions, how can financial institutions retain effective control over their behaviour?

BENCHMARK ANALYTICS | FINANCIAL MARKETS The machines are doing the trading ASIC estimates of algorithmic trading in major Australian markets

SPI 200 futures Hedging Australian share-market exposure 94%

Australian listed equities Buying and selling listed company shares 85%

Three-year Treasury bond futures Trading short-term government bond exposure 46%

ASIC's regulatory response

2028 New rules commence 18 March; 18-month transition ~60% Less guidance proposed Relevant securities guidance

Sources: ASIC, August 2025 trading estimates and September 2026 market integrity rule amendments. Percentages are estimates of algorithmic trading, not AI trading. The guidance reduction remains a proposal.