ABC host Alan Kohler warns government will be forced to introduce another tax in Australia
AI, designed to replace human labour and boost company profits, threatens to erode the federal government's biggest source of revenue, personal income taxes, Kohler warned.
This would see the government have less money to spend on essential services like welfare, transport infrastructure and defence.
Personal income taxes make up a majority of federal government revenue and Treasury is expecting to be even more reliant on this revenue source into the late 2020s, even as technology replaces jobs.
To solve this problem, Kohler has suggested a new tax on AI.
'While the taxes on human labour are increasing, the spending on artificial intelligence designed to replace human labour is going through the roof and so are the profits,' he said. 'And why are the companies going for AI? Well, largely to replace staff.'
Geopolitical political uncertainty and an ageing population also gives the government less scope to cut spending to cope with a future plunge in revenue from personal income taxes - leaving a tax on AI as the only option.
'Good luck cutting spending to match the decline in personal income tax revenue as artificial intelligence starts replacing taxpaying human workers, governments will either have to tax the profits from robots and AI or tax wealth,' he said.
The federal government is expecting to collect $349.7billion from income taxes in 2025-26, which would make up 51.7 per cent of the Commonwealth's $676.1billion in revenue.
By the 2028-29 financial year, Treasury is expecting personal income taxes to make up 54 per cent of revenue as receipts from individuals soared to $420.3billion from a total collection pool of $778.3billion.
The March Budget papers expected this to occur even as technology giants like Google, Microsoft, Amazon, Apple and OpenAI spent even more on artificial intelligence large language models.
Global artificial intelligence investment hit $200billion in 2024 and Forbes is expecting it to approach $400billion this year, in Australian dollar terms.
Kohler noted the federal government was instead focused on applying a 15 per cent tax on unrealised gains on superannuation balances of more than $3million, without indexing it for inflation.
He slammed the idea of taxing retirement savings without indexing it for inflation, after AMP forecast the tax would affect the average, 22-year-old worker in four decades time.
'So, it's not just a wealth tax, it also brings bracket creep to super,' he said.
'And it may not be the last tax on wealth either.'
With AI threatening to replace jobs, increasing taxes on the highest 0.5 per cent of superannuation balances may do little to compensate for the collapse in personal income tax revenue.
'And the tax on high super balances is just a toe in that water,' he said.
The chief executives of the Commonwealth Bank' and Telsta - Matt Comyn and Vicki Brady - told last week's Australian Financial Review AI Summit that artificial intelligence was advancing at a faster pace than many people anticipated.
'Everyone talks about Moore's law, that computer power doubles every two years. The capability of these agents is doubling every seven months,' Ms Brady said.
Mr Comyn predicted AI would take away customer service jobs in banking.
'Whereas in other areas … around customer service, where there is greater automation, I think some of those roles will be challenged,' he said.
White collar jobs are most at risk of being replaced by AI with the likes of tax and payroll accountants and banking staff in danger, a Mandala Partners report predicted in 2023.
A tax on AI could potentially be used to fund a universal basic income, where everyone gets a guaranteed government payment without a means test.
Basic Income Australia pitched this idea to a Senate committee on adopting artificial intelligence, but the inquiry last year declined to recommend that policy.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Geeky Gadgets
39 minutes ago
- Geeky Gadgets
3 AI Concepts Everyone Misunderstands Simplified
What if the secret to unlocking Artificial Intelligences AI's full potential wasn't about mastering complex algorithms, but simply understanding a few misunderstood ideas? Artificial intelligence is everywhere—from chatbots to content creation—but the jargon surrounding it often feels like a barrier rather than a bridge. Terms like 'tokenizable data' or 'jagged intelligence' might sound intimidating, but they hold the keys to making AI work smarter for you. Here's the truth: you don't need to be a data scientist to grasp these concepts. You just need a clear explanation of what they mean and how they impact your interactions with AI. In this overview, Nate Jones demystifies three of the most misunderstood concepts in AI, breaking them down into practical insights you can apply immediately. You'll learn why clean, structured data is the unsung hero of AI performance, how to navigate the uneven strengths of 'jagged intelligence,' and when to use detailed prompts versus casual chats for better results. These aren't just abstract ideas—they're tools to help you collaborate more effectively with AI and avoid common frustrations. By the end, you might just see AI not as a mysterious black box, but as a partner you can guide and optimize. After all, understanding is the first step to mastery. Mastering AI Fundamentals What is Tokenizable Data? Tokenization is a fundamental process in AI systems, where information is broken down into smaller, manageable units called tokens. These tokens allow AI to process and analyze data more efficiently, making tokenization a cornerstone of how AI understands and interacts with information. Text-based formats, such as plain text files or Word documents, are inherently easier for AI to tokenize, making them ideal for tasks like summarization or content generation. On the other hand, more complex formats, such as spreadsheets or vast datasets like data lakes, can pose challenges due to their intricate structures and size. The quality of your data plays a pivotal role in determining how effectively AI performs. Clean, well-organized data enables the system to focus on the task at hand rather than wasting resources interpreting messy or unstructured inputs. For instance, if you want AI to summarize a report, providing a plain text version will yield far better results than submitting a cluttered PDF filled with extraneous formatting. By understanding tokenization, you can prepare your data in ways that align with AI's strengths, making sure smoother processing and more accurate outputs. Jagged Intelligence: Why AI Excels in Some Areas and Struggles in Others AI is not a uniformly intelligent system; its capabilities are uneven, a phenomenon often referred to as jagged intelligence. This means that while AI can excel in certain areas, it may struggle in others. For example, AI is highly effective at tasks like generating coherent text, identifying patterns in large datasets, or performing repetitive, rule-based operations. However, it often falters when faced with challenges requiring long-term memory, nuanced reasoning, or even basic arithmetic under specific conditions. To work effectively with AI, it is crucial to understand and navigate these inconsistencies. By focusing on the areas where AI excels and compensating for its weaknesses, you can achieve better results. For instance, high-quality prompts and intentional input can help bridge performance gaps. If you're using AI for creative writing, a clear and detailed prompt will guide the system to produce more relevant and coherent outputs. Conversely, expecting AI to handle complex, multi-step reasoning without guidance will likely lead to disappointing results. Recognizing the concept of jagged intelligence allows you to set realistic expectations and design workflows that capitalize on AI's strengths. This approach not only improves the quality of the outcomes but also minimizes frustration when working with these systems. AI Jargon Demystified Watch this video on YouTube. Browse through more resources below from our in-depth content covering more areas on Artificial Intelligence. Big Prompts vs. Casual Chats: Tailoring Your Approach The way you interact with AI—whether through detailed, big prompts or shorter, iterative exchanges—can significantly influence the quality of its outputs. Each approach has its advantages, depending on the nature of the task at hand. Big Prompts: These are ideal for tasks that require structured, context-rich responses. For example, if you're asking AI to draft a business proposal, a comprehensive prompt that outlines your objectives, target audience, and key points will produce the best results. Big prompts provide the AI with a clear framework, reducing ambiguity and improving the quality of the output. These are ideal for tasks that require structured, context-rich responses. For example, if you're asking AI to draft a business proposal, a comprehensive prompt that outlines your objectives, target audience, and key points will produce the best results. Big prompts provide the AI with a clear framework, reducing ambiguity and improving the quality of the output. Casual Chats: Shorter, iterative exchanges are better suited for exploratory tasks like brainstorming or generating ideas. This approach allows you to refine the AI's output step by step, adapting your input based on its responses. For instance, when brainstorming marketing slogans, you might start with a broad idea and gradually narrow your focus through a series of back-and-forth interactions. Choosing between big prompts and casual chats depends on your specific goals. Structured tasks benefit from detailed instructions, while exploratory tasks thrive on flexibility and iteration. By aligning your interaction style with your objectives, you can maximize the AI's potential and achieve more effective results. Key Takeaways for Effective AI Collaboration To collaborate effectively with AI, it is essential to understand its core mechanisms and limitations. Here are the main points to keep in mind: Tokenizable Data: AI processes clean, structured data more efficiently. Prepare your inputs in simple formats like plain text to ensure optimal results. AI processes clean, structured data more efficiently. Prepare your inputs in simple formats like plain text to ensure optimal results. Jagged Intelligence: AI excels in specific areas but struggles in others. Focus on using its strengths and use high-quality prompts to address its weaknesses. AI excels in specific areas but struggles in others. Focus on using its strengths and use high-quality prompts to address its weaknesses. Big Prompts vs. Casual Chats: Tailor your approach based on the task. Use detailed prompts for structured outputs and iterative exchanges for exploratory work. By mastering these concepts, you can unlock the full potential of AI systems. Whether you're analyzing data, generating content, or brainstorming ideas, a deeper understanding of tokenization, jagged intelligence, and prompt design will empower you to achieve better outcomes. These principles are not just theoretical—they are practical tools that can help you navigate the complexities of AI and make the most of its capabilities. Media Credit: AI News & Strategy Daily | Nate B Jones Filed Under: AI, Top News Latest Geeky Gadgets Deals Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, Geeky Gadgets may earn an affiliate commission. Learn about our Disclosure Policy.
Reuters
2 hours ago
- Reuters
Australia's green bank makes record $2.3 billion in clean energy investments
SYDNEY, July 28 (Reuters) - Australia's green bank on Monday said it invested a record A$3.5 billion ($2.29 billion) in clean energy projects and grid infrastructure last financial year to accelerate the country's transition from fossil fuels to renewables. Australia's coal power generation makes it one of the highest polluting countries per capita in the world. It plans to shut all coal stations by 2038 and has pledged to achieve a target of 82% renewable generation by 2030. But analysts have called for increased investment to meet this goal, achieve net zero emissions by 2050 and ensure energy security. Wood Mackenzie projects Australia is on track to hit only 58% renewable energy generation by the end of the decade. The Clean Energy Finance Corporation, Australia's A$32.5 billion green bank, has a mandate to 'facilitate increased flows of finance into the clean energy sector and to facilitate the achievement of Australia's greenhouse gas emissions reduction targets'. During the 2024-25 financial year, it committed A$4.7 billion, including A$3.5 billion to renewable energy projects and grid infrastructure. The CEFC said it was a 'record' level of investment and more than double the amount of the previous fiscal year. The largest transaction was A$2.8 billion towards a programme to improve Australia's national power grid, including A$2.1 billion for construction of a new electricity transmission link on the country's east coast. The CEFC said it was ramping up investment in clean energy to help Australia meet its renewable energy and emissions targets. 'Australia requires ongoing investment in renewables and long-duration storage, clean energy affordability for consumers and measures to cut emissions,' CEO Ian Learmonth said. 'Investment activity at this scale promises substantial economic and local employment benefits across Australia, strengthening our economy for a net zero future while making critical progress towards decarbonisation.' ($1 = 1.5267 Australian dollars)
Reuters
3 hours ago
- Reuters
WiseTech Global taps insider Zubin Appoo as new CEO, succeeding founder Richard White
July 28 (Reuters) - Australian logistics software maker WiseTech Global ( opens new tab on Monday said its chief of staff, Zubin Appoo, will take on the role of CEO effective immediately, succeeding billionaire founder Richard White. White, the company's long-serving chief executive, put down his papers last October, following a flurry of controversies, including media reports of allegations about his personal life. White's exit comes amid a broader crisis marked by boardroom exodus, corporate governance concerns, and share volatility that led to Australia's largest pension fund, AustralianSuper, selling its stake in WiseTech Global. Appoo previously spent 14 years at WiseTech from 2004 to 2018 as head of innovation and technology, and was part of the team that developed WiseTech's flagship CargoWise platform. He returned to the company earlier this year as the deputy chief innovation officer, reporting directly to White. "Zubin Appoo's appointment is a much-needed stabilising force" for WiseTech, said Josh Gilbert, market analyst with eToro. "His elevation to CEO allows the business to draw a line under recent boardroom drama and provides a clearer leadership structure moving forward." Shares of the company ended 0.3% higher at A$120.5. Andrew Cartledge, who served as the logistics tech firm's finance chief for nearly a decade, took over the interim CEO duties in October 2024. The company reiterated that Cartledge will retire at the end of the year as previously stated, but did not specify what role he will take on till his retirement. Last year in October, the Australian Financial Review and other media outlets reported that a woman who had had a sexual relationship with White made numerous allegations against him in late 2020, including claims of inappropriate behaviour. Initial findings of an external governance review started by the company largely cleared White of wrongdoing, though it acknowledged that his management style might be perceived by some employees as intimidating. The billionaire made a surprise return to the firm's top brass in February, appointed as its executive chair.
