After the unsatisfying launch of Bard in February 2023, Google returned with the improved version of its artificial intelligence (AI) in the Google I/O event on May 10th, 2023. The first launch was criticized as a hurried effort, a panicked response to the launch of OpenAI’s ChatGPT which is backed by Microsoft, one of Google’s main competitors. OpenAI’s artificial intelligence was able to provide a response unlike any other and planned to be integrated to Microsoft’s search engine, Bing. Though receptions of the new version of Google’s AI programs were mixed, everyone will agree that Google is giving their extra attention and effort to not being left behind in this major trend. AI is being all the main topic of discussion all around lately, as the improvement and development of it was way quicker than anyone imagined. Though flaws still exists, current AI are able to surprise us with their capabilities. For example, ChatGPT passed law exams in four courses at the University of Minnesota and another exam at University of Pennsylvania’s Wharton School of Business, according to professors at the schools, though not with particularly high marks.
AI is rapidly reshaping industries and economies worldwide, revolutionizing the way we work and interact with technology. While AI brings immense potential and benefits, there are concerns that further development and implementation of AI may resulted in massive layoffs by as it replaces human workforce and eliminate businesses that could not invest in the technology. According to PwC research, by the mid-2030s, one-third of all employment will be at risk of being automated. The workforce segment most likely to be affected will be individuals with a low level of education. COVID-19 managed to further boost the effort in the sector by forcing companies to minimize infections in the workplace while keeping operational expenses low. How? To replace humans with machinery. We can see several examples:
With the furtherly improved capabilities of AI, it is no longer far to imagine a future without work. a study co-authored by an MIT professor puts firm numbers on the trend, finding a very real impact. MIT economist Daron Acemoglu finds fairly major negative employment effects on implementation of machinery on the workplace. From 1990 to 2007, the study shows, adding one additional robot per 1,000 workers reduced the national employment-to-population ratio by about 0.2 percent in the US. The study also finds negative wage effects, that workers are losing in terms of real wages in more affected areas, because robots are pretty good at competing against them.
Numbers issued by PwC on employment shares and the proportion of jobs at high risk of automation by the early 2030s also showed similar story to the research in the US, AI and machines are able to replace human labor.
Source: PwC
Table 1: Share of Employment and Risk of Job Automation by Industry
|
Industry |
Share of Employment |
Risk of Job Automation |
|
Wholesale and retail trade |
14.80% |
44% |
|
Manufacturing |
7.60% |
46.4% |
|
Administrative and support services |
8.40% |
37.4% |
|
Transportation and storage |
4.90% |
56.4% |
|
Professional, scientific and technical |
8.80% |
25.6% |
|
Human health and social work |
12.40% |
17% |
|
Accommodation and food services |
6.70% |
25.5% |
|
Construction |
6.40% |
23.7% |
|
Public administration and defense |
4.30% |
32.1% |
|
Information and communication |
4.10% |
27.3% |
|
Financial and insurance |
3.20% |
32.2% |
|
Education |
8.70% |
8.5% |
|
Arts and entertainment |
2.90% |
22.3% |
|
Other services |
2.70% |
18.6% |
|
Real estate |
1.70% |
28.2% |
|
Water, sewage and waste management |
0.60% |
62.6% |
|
Agriculture, forestry and fishing |
1.10% |
18.7% |
|
Electricity and gas supply |
0.40% |
31.8% |
|
Mining and quarrying |
0.20% |
23.1% |
|
Domestic personnel and self-subsistence |
0.30% |
8.1% |
|
Total/Average for all sectors |
100% |
30% |
But we must not only look from one side. While AI may replace certain tasks, it also creates new job roles and demands a different skill set. Future of Jobs Report 2023, by the World Economic Forum stated that 74.9% of the organizations it surveyed are likely or highly likely to adopt AI over the next 5 years.
Exhibit 1: Future of Jobs Report 2023
Source: Future of Jobs Report 2023, World Economic Forum
But in total, or net effect, 25.6% would expect that adopting AI will create more jobs.
Exhibit 2: Expected impact of technology adoption on jobs, 2023-2027
Share of organizations surveyed that expect each technology to create or displace jobs, ordered by the job creation net effect. The shares of organizations which expect the impact of adopting these technologies to be neutral are not plotted.
The 2020 version of the report expect that by 2025, new jobs in AI related field will be in demand as the adoption in the sector.
Exhibit 3: Job Landscape
By 2025, new jobs will emerge and others will be displaced by a shift in the division of labour between humans and machines, affecting:
Source: Future of Jobs Report 2020, World Economic Forum
While the forecast may doesn’t look as worrying as we previously thought, we still need to make adjustment in the workforce in order to properly adapt to the shift of need in the job market. Luckily – or not –AI, ChatGPT, itself provided us with a paper on jobs that AI will replace to know which area that we should focus or shift upon.
The 35-page research paper on GPTs’ labor market impact was published on March 21, 2023. The paper finds that 80% of US jobs will be exposed to GPT. Replaced or exposed means the likelihood of a company or business using GPT to perform 50% of the task that a human can do.
The paper indicates that most occupations that require science and critical thinking skills are less likely to be impacted by GPT. However, writing and programming have a higher exposure which means occupations that need the two skills can rely on GPT to perform 50% of tasks done by a human. The paper also notes that employees with College, Bachelor, and Master’s degrees are more exposed to GPT than those with informal education. However, jobs that require long training are less exposed.
Table 2: Occupation with exposure to the implementation of ChatGPT
But on the other side of the spectrum, the paper also found that jobs that requires manual labor are not likely to be replaced with AI.
Table 3: Occupations Without Any Exposed Tasks
So it’s either developing skills of the workforce towards science and critical thinking or training on the abilities required on the manual labor intensive jobs.
Indonesia was considered to be one of the least prepared countries according to an article from Forbes in 2018 on job loss from automation, robotics and AI. Indonesia ranked last out of 25 countries, measured by the national R&D budget as % of GDP by only spending 0.08%. Indonesia also the last when compared to other countries in the region such as Vietnam (0.37%) and Malaysia (1.30%).
Exhibit 4: Gross expenditure on R&D as % of GDP, 2016
Source: UNESCO Institute for Statistics
Indonesia should not stay in comfort on this position to maintain employment of its workforce as companies in the economy considered increased adoption of new and frontier technologies to be their second highest priorities, after the green transition of their businesses, based on the World Economic Forum’s report. This signals that implementation of AI will be included and most certainly affect the workforce demand by reducing those that can be replaced and increased demand to support the smooth transition in using the technology.
Exhibit 5: Global trends and their impact on job creation
Trends most most likely to drive industry transformation and their expected impact job creation, ordered by net effect (share of organizations surveyed).
On the other hand, there are opportunities that Indonesia and its workforce must prepare and get from this implementation of technology. In net, 54% of the companies or organization in Indonesia believed that implementation of AI will be a job creator, demanding more people to work for the implementing companies to be able to use the technology to their fullest potential. With 48% of employment in Indonesia is considered vulnerable to disruptions by the World Economic Forum, the implementation of technology such as AI will be most likely affecting this portion of the workforce negatively. Efforts on reskilling and upskilling will be required from the policy maker, businesses, and individuals to prepare for the coming wave of AI and other automation technologies.
Exhibit 6: Technologies and their impact on job creation
Technologies most most likely to drive industry transformation and their expected impact job creation, ordered by net effect (share of organizations surveyed)
Artificial intelligence is a transformative force that profoundly impacts the economy and the workforce. While it presents challenges such as job displacement, it also creates opportunities for innovation, productivity growth, and new job roles. To harness the benefits of AI, policymakers, businesses, and individuals must proactively adapt, invest in reskilling and upskilling, and foster a collaborative relationship between humans and AI. By embracing AI responsibly, we can shape a future where humans and intelligent machines coexist, driving economic progress and societal well-being.
But for those who feel that their current jobs involve a lot of sitting in front of the computer and manual data input or calculation, upskilling or reskilling is needed. If not, as AI won’t affect much of the manual labor jobs, maybe we should consider working in a farm?
