MG RESEARCH at Assumption University: Flying Geese, Rising Dragons: Revisiting Wad's Malaysia–Thailand Automobile Divergence in the Chinese EV Era | Academic Research Blog | November 8, 2026

Anchor paper: Peter Wad (2009), "The automobile industry of Southeast Asia: Malaysia and Thailand," Journal of the Asia Pacific Economy 14(2): 172–193.

https://www.researchgate.net/profile/Peter-Wad/publication/232937524_The_automobile_industry_of_Southeast_Asia_Malaysia_and_Thailand/links/54db870e0cf2a7769d91fba3/The-automobile-industry-of-Southeast-Asia-Malaysia-and-Thailand.pdf

Abstract

In 2009 Peter Wad argued that Southeast Asia's two largest car-making economies had run a natural experiment. Thailand opened its doors to foreign multinationals (FDI-led), while Malaysia built a state-sponsored national champion, Proton. The 1997–98 East Asian financial crisis, he claimed, was the turning point: it forced multinationals in Thailand to export, while Malaysia's national brands lacked the technology, brands and distribution networks to follow. His verdict was that Thailand was a success story and that it was "last call" for Malaysia's national project, which would have to either find export niches or re-link with a foreign lead firm.

Overview of the Paper's Abstract and Core Argument

The reviewed article investigates the evolution and restructuring of the automotive industries in Southeast Asia—specifically focusing on Malaysia and Thailand—a decade after the 1997–1998 East Asian financial crisis and amidst the subsequent global economic downturn [1]. The paper evaluates how these two nations, which started with similar import-substitution policies in the late 1960s, diverged into contrasting strategic pathways: Malaysia pursued a state-driven national-champion policy, whereas Thailand embraced an open-door policy centered on foreign direct investment (FDI) [1], [2].

Key Comparative Findings and Strategic Divergence

• Thailand's Export-Led Success: Thailand emerged as a major regional and global success story in the twenty-first century by specializing in commercial vehicles—particularly 1-ton pickup trucks—integrated through the global value chains of Japanese and American multinational corporations (MNCs) [1], [2]. The East Asian financial crisis acted as a turning point, forcing MNCs in Thailand to pivot decisively toward export-oriented manufacturing [2].
• Malaysia's Stalling National Project: Conversely, Malaysia's national automobile project (epitomized by Proton) struggled with international competitiveness, weak technological capabilities, and inadequate distributive networks [2]. Confined heavily to protected domestic markets and targeting less optimal export segments (such as the United Kingdom), the national policy faced severe strains when local demand slumped [2], [3].
• Strategic Options for Malaysia: The study posits that Malaysia's national automobile initiative faces an urgent crossroads: it must either innovate export-driven brands targeting less competitive markets (such as Islamic countries) or re-link with global MNCs [1]. However, re-linking requires relinquishing local management control—a transition successfully achieved by Perodua but unsuccessfully attempted by Proton [1], [4].

Limitations in the Automotive Component Sector

• Marginalization of Local Suppliers: Both the Thai and Malaysian automotive sectors failed to foster an internationally competitive, locally controlled small and medium enterprise (SME) component sector [5].
• Thai Supplier Suppression: In Thailand, local auto component suppliers were largely marginalized or pushed out by dominant foreign—especially Japanese—suppliers [5].
• Malaysian Vendor Weaknesses: Although Malaysia's national automotive policy actively promoted local auto component vendors under ethno-economic transformation frameworks, these local suppliers suffered from low technological capabilities, poor financial performance indicators, and an inability to export effectively [5].

Socio-Economic Implications for Autoworkers

• Erosion of Relative Wages: Historically, Malaysian autoworkers enjoyed higher relative remuneration levels compared to counterparts in neighboring countries [5]. However, these wage advantages and working conditions have come under severe pressure due to the relative decline and restructuring of Malaysia's domestic automobile sector [5].
• Recovery in Thailand: Meanwhile, Thai autoworkers experienced high relative wage increases post-crisis, successfully regaining and surpassing pre-crisis real wage levels as industry production and employment expanded [6

Contributions of the Paper

This paper makes several key contributions to the economic and sociological understanding of industrial development, comparative political economy, and automotive manufacturing in Southeast Asia [1]. By analyzing the divergent trajectories of Malaysia and Thailand a decade after the 1997–1998 East Asian financial crisis, the article provides valuable empirical insights and theoretical frameworks regarding national industrial policies and global value chains [2].

Comparative Analysis of Divergent Industrial Strategies

• Contrasting National Pathways: The paper contrasts Malaysia's state-driven national-champion policy (exemplified by Proton) with Thailand's open-door, foreign direct investment (FDI)-oriented strategy [1]. It demonstrates how these historical choices made since the early 1980s led to drastically different outcomes when subjected to economic shocks [1], [3].
• Impact of the East Asian Financial Crisis: The study identifies the 1997–1998 financial crisis as a critical turning point [2]. It explains how the crisis forced multinational corporations in Thailand to pivot successfully toward export-oriented manufacturing, whereas Malaysia's national automotive projects stalled due to weak technological capabilities, restrictive branding, and inadequate international distribution networks [2].

Evaluation of Component Sectors and Labour Dynamics

• Supplier Development Assessment: The research evaluates the peripheral automotive component and parts sectors, documenting how both nations struggled to foster internationally competitive small and medium enterprise (SME) suppliers [4]. It highlights how Thai local suppliers were largely marginalized by dominant foreign entities, while Malaysian vendors underperformed on financial and technological indicators [4].
• Socio-Economic Implications for Autoworkers: The paper addresses the broader socio-economic impacts on autoworkers, tracing how real wages and employment conditions shifted in tandem with industrial performance—noting that Thai autoworkers surpassed pre-crisis wage levels by 2005, whereas Malaysian workers faced growing relative pressures [5].

Strategic Implications and Policy Lessons

• Future Options for National Projects: The article offers concrete strategic alternatives for lagging national initiatives like Malaysia's Proton, suggesting either a pivot toward less competitive export markets (such as Islamic nations) or a re-linking with multinational corporations by relinquishing local management control [1].
• Integration within Regional and Global Value Chains: By framing its analysis between global/regional value chains and domestic institutional arrangements, the paper contributes to the broader literature on how developing nations can navigate integration, trade liberalization, and regional free trade areas like the ASEAN Free Trade Area [3].

Overview of Thailand's Successful Automotive Strategies

Thailand implemented an open-door, foreign direct investment (FDI)-oriented automotive strategy that successfully transformed its domestic manufacturing sector into a regional and global export hub [1]. Rather than pursuing state-led national champions, Thailand relied on multinational corporations (MNCs) and market-driven integration within global value chains [2].

Core Strategic Pillars and Turning Points

• Open-Door Policy and FDI Liberalization: Thailand opened its automobile industry to foreign multinational enterprises starting in the early 1990s, culminating in the complete removal of local content requirements by 2000 [3]. This welcoming institutional environment allowed fully owned subsidiaries of global automakers to establish roots [4].
• Strategic Specialization in Commercial Vehicles: Taking advantage of domestic demand patterns and tax incentives, Thailand specialized in the production and export of 1-ton pickup trucks [5]. This targeted niche successfully established Thailand as the primary global hub for pickup manufacturing outside the United States [6].
• Post-Crisis Export-Led Pivot: The 1997–1998 East Asian financial crisis acted as a vital turning point [7]. Faced with a slumped domestic market and heavy sunk costs, MNC affiliates in Thailand pivoted decisively toward export-oriented manufacturing, leveraging depreciated currencies and liberal trade policies [8].
• Enhanced Public-Private Coordination: Under the Thaksin administration in the early twenty-first century, the government launched the "Detroit of the East" initiative [9]. This plan further aligned industry support with export performance, established the Thai Automotive Institute, and fostered robust public-private collaboration through "Thailand Inc" [10].

Synthesis of Outcomes

Through these combined measures, Thailand successfully integrated into the value chains of Japanese and American MNCs, generating substantial trade surpluses, employment, business turnover, and public revenues despite marginalizing local component suppliers in favor of foreign multinational networks [11].

This post tests that verdict against 2026 evidence from three sources: Bloomberg's news and Bloomberg NEF coverage of the ASEAN auto market, Moody's commentary on the auto sector, and the corporate record of two Shanghai Stock Exchange-listed groups, SAIC Motor and Great Wall Motor, that now operate plants in Thailand. My conclusion is that Wad's mechanism (lead-firm strategy plus local capability, filtered through domestic political economy) survives and is arguably strengthened. His cast of characters does not. The lead geese are no longer Japanese, the Thai model is now defended with the same protectionist tools Malaysia once used, and Malaysia's champion has survived by doing the thing Wad said it would have to do: partnering with a foreign firm.

1. Why revisit a 2009 paper now?

Wad's paper was written at the trough of the global financial crisis, using data that stop in 2007. Three things have changed since. First, the industry's centre of gravity has shifted from combustion engines to battery-electric vehicles (BEVs), where China is the dominant producer. Bloomberg NEF's 2026 Electric Vehicle Outlook reports that China accounted for 63% of global electric car sales in 2025, and that Thailand's EV share of new car sales reached 27% that year, with Chinese brands making up 88% of EVs sold there.

Second, the regional lead firms have changed. In Wad's account the Japanese keiretsu were the "lead geese" of a regionalised production structure, and Western makers had largely abandoned the region. Today Chinese groups are building plants in Thailand, often in factories vacated by Western makers.

Third, policy has swung back toward protection. Both countries now use tax design to steer investment toward local assembly, which complicates Wad's finding that industrial policy was not the decisive variable.

The exercise matters beyond autos. If a framework built on 1997–2007 data explains 2026 outcomes, it is a durable tool for thinking about how developing economies capture value in globally governed production networks. If it fails, we learn which of its assumptions were period-specific.

2. Wad's argument in brief

2.1 Framework

Wad combines a historical-institutionalist perspective with global value chain (GVC) analysis. Growth and net exports of vehicles and parts are the outcomes to explain. They are conditioned by four things: the national automobile regime (policy), the strategies and capabilities of lead firms (OEMs) and first-tier suppliers, the response of local assemblers and suppliers, and the wider political economy (state capacity, national innovation system, industrial relations). The automobile industry is classically a producer-driven value chain, though Wad notes that since the 1970s outsourcing has made governance more relational.

He also draws on the flying-geese tradition: Japanese assemblers and their suppliers moved across Southeast Asia as a regional keiretsu, with local firms chasing the flock. That structure proved resilient against both indigenous national projects and the return of Western makers in the 1990s.

2.2 The pessimistic prior

Wad opens with the GVC-school consensus that national car projects in small and mid-sized developing economies are futile: entry barriers are high, scale economies are demanding, and the investment is better spent elsewhere. He cites automotive analysts Maxton and Wormald, who judged that peripheral Asian players such as Proton were unlikely to survive a coming restructuring of the global industry. The paper's task is to ask whether Thailand and Malaysia confirm that prior.

2.3 Core claims

1. Divergent regimes. Both countries began with import-substitution in the late 1960s. Malaysia launched its national car programme in the 1980s through a majority-controlled joint venture with Mitsubishi. Thailand liberalised in the early 1990s and abolished local-content rules in 2000.
2. The crisis as turning point. Before 1997 liberal Thailand did not outperform nationalist Malaysia. After it, multinationals with sunk costs in Thailand had no choice but to export, while Proton and Perodua, built for a protected home market, could not.
3. Policy is not the whole story. Wad's "sub-conclusion" is that industrial policy was neither the decisive nor the only factor. The drivers sit at the micro level of firm capabilities and strategies.
4. A shared failure. Neither country built an internationally competitive, locally controlled supplier sector. Japanese and other foreign suppliers marginalised Thai firms, and Malaysia's ethno-economic policy (the NEP) sustained local vendors that remained uncompetitive.
5. Political economy. Malaysia's Bumiputera-oriented policy sidelined existing Chinese-Malaysian industrial skills. Thailand's more assimilated business elite pursued a market and MNC-driven path instead.
6. The verdict. Thailand is a success via the value chains of Japanese and American multinationals. Malaysia faces "last call": design exportable brands for less demanding markets (Wad names Islamic countries) or re-link with a multinational, probably surrendering management control.

3. What the 2000–2007 data showed

The empirical heart of the paper is a set of production and trade tables. A compact version:

Indicator (Wad's tables)	Malaysia	Thailand
Motor vehicle production, 2000	284,600	325,888
Motor vehicle production, 2007	413,440	1,238,460
Change 2000–2007	+45%	+280%
Vehicle trade balance 2000 (US$ m)	−1,149.4	+1,178.0
Vehicle trade balance 2006 (US$ m)	−1,511.0	+6,089.2
Parts trade balance 2006 (US$ m)	−561.0	−58.4
Domestic vehicle sales, 2007	487,176	631,250

A small note on the data: the text of the paper describes Thai growth as 285%, while Table 1 computes 280% (912,572 additional units on a base of 325,888, which is 280%). The difference is immaterial to the argument, but it is a reminder that Wad leans on OICA estimates, some of which he flags as approximate for Indonesia, the Philippines and Vietnam.

Three patterns stand out. Thailand's output nearly quadrupled and the industry specialised in one-tonne pickups, which accounted for more than 60% of Thai vehicle output in the 1990s and then drove exports. Thailand's vehicle trade surplus grew roughly fivefold between 2000 and 2006, while Malaysia's deficit widened. And yet Thailand still ran a parts deficit, a flag that high export value did not automatically mean deep local value-added.

On market structure, Wad shows Japanese makers controlling 81% of Thai domestic sales in 2001, and Proton's market share falling from a 74% peak in 1993 to nearly 20% by 2007 as Toyota and Honda pushed cheaper, high-quality models and Perodua (with Daihatsu technology and, later, Daihatsu majority ownership) became the best-selling Malaysian make.

4. Why the paths diverged: Wad's mechanism

The central causal chain is sunk cost plus crisis plus currency. When the baht collapsed and domestic demand cratered in 1997–98, multinational affiliates in Thailand faced bankruptcy. Their plants were already built, the local-content rules were being dropped, and a depreciated currency, a liberal trade regime and weak, fragmented unions made Thailand a cost-effective export base. The multinationals chose to turn Thailand into a hub for commercial vehicles. In Malaysia, by contrast, Proton and Perodua had a captive home market and weak brands, technology and distribution abroad. Their early export push, aimed heavily at the UK, targeted exactly the wrong kind of market: mature, quality-sensitive and intensely competitive.

Wad stresses what the Malaysian state did right (it bought design capability by acquiring Lotus and developed in-house engines) and what it failed to do. It did not mobilise existing industrial capabilities, in particular Chinese-Malaysian entrepreneurs. It did not tie support to export performance, as Korea did. And it did not build a national innovation system around the programme. He attributes all of this to the political economy of the New Economic Policy.

The framework has two virtues worth keeping. It is falsifiable: it predicts that export orientation, supplier capability and lead-firm linkages should explain outcomes better than headline policy. And it is multi-level, forcing us to ask who the lead firm is, who the local firms are, and what the state is trying to do. Both virtues are useful when we turn to 2026.

5. Cross-examination I: Bloomberg and Bloomberg NEF on Thailand's EV shock

A note on sources. Bloomberg Intelligence's sector research sits behind the Bloomberg Terminal and could not be accessed for this post. I have instead used Bloomberg News reporting and Bloomberg NEF's publicly available outlook, which draw on the same market. Where I cite figures, they come from those public pieces.

5.1 Thailand is no longer simply a Japanese-led export success

The 2009 paper's headline was Thai export success through Japanese and American value chains. The 2026 record is more unsettled. Thai output was 1,455,569 vehicles in 2025, a 0.9% dip, according to the Federation of Thai Industries (FTI), which targets about 1.5 million for 2026. That is only about 17% above the 2007 figure in Wad's Table 1, after a 280% surge in the preceding seven years. In the first five months of 2026 output was down 1.1% year on year, and September production fell 6.1% to 120,319 units, with the FTI blaming late-September to mid-October flooding that disrupted component deliveries. Earlier in the year the FTI and trade analysts pointed to Middle East conflict disrupting export logistics. The honest reading is that the Thai industry has stopped compounding and is in a transition with noisy monthly data.

5.2 The import surge is the new "1997"

Where Wad saw the Asian crisis forcing a pivot to exports, 2024–26 looks like a pivot to localisation under competitive pressure, with Chinese BEV imports as the shock. Bloomberg reported that in the first nine months of 2026 Thailand imported US$2.37 billion of electric passenger cars, 96% from China. September imports rose 138% year on year to US$421 million, with China supplying almost 99%. Importers were front-loading shipments ahead of a higher excise tax that takes effect on 1 January. According to Bloomberg's 6 November report, the government has announced a 35% excise tax on imported EVs, in a three-tier structure that taxes locally made EVs least, vehicles with partly imported components in the middle, and fully imported models most. Bangkok wants locally made cars to account for 70% of supply within two years, more than double their share in the first nine months of 2026. The FTI's automotive club expects EV sales of about 200,000 in 2026.

The response is just what Wad's sunk-cost logic would predict. A Thai distributor of XPeng vehicles told Bloomberg it was studying a local factory, and Omoda & Jaecoo said it would raise local output and component use. Tax design is pulling Chinese lead firms into Thai plants.

5.3 What this does to Wad's "policy is not decisive" claim

Wad concluded that Thai policy did not play the decisive role because the Thaksin-era "Detroit of the East" plan produced little in the way of heavy state investment. Today, tiered excise rates, local-content thresholds and battery-localisation rules are central to firm decisions. Counterpoint-style industry commentary describes Thailand's consistent EV policy framework as the reason it attracts battery and supply-chain investment, and Vietnamese and Thai press reports cite roughly US$4.1 billion in EV supply-chain pledges. I read this as a partial refutation: policy mattered less in 1997–2007 because the lead firms' strategic needs and the policy direction coincided. In 2026 they do not coincide, and policy is doing real work.

6. Cross-examination II: Malaysia's champion did not die

This is where the 2026 record most surprises a 2009 reader.

6.1 The national brands have reclaimed share

Wad reported Proton falling to roughly 20% of the market by 2007. In 2026 the picture reversed. Hong Leong Investment Bank reported total industry volume (TIV) up 4.1% to 604,700 units in the first nine months, with Proton sales up 39.2%. Proton and Perodua together held 66.3% of the market, up from 62.1% in 2025, and Perodua alone held 40.9%. Trade-press registration data (a different denominator) put Proton at 23.9% to September, its highest since 2011, and put the combined national share at 62.5%, the highest since 2003. The two sources differ on the exact percentages, but they agree on the direction. Analysts at Business Today noted that the combined share was below 50% before 2019.

Proton's recovery has an identifiable cause: its partnership with China's Geely. Paultan, a Malaysian auto publication, notes that Proton's first-half 2018 share was just 9.1% before the partnership began to bear fruit, and that its half-year volume is now about 3.6 times that low. Its best-selling models include Geely-based vehicles (the X50 and S70, which are rebadged Geely Coolray and Emgrand), and its e.MAS 5 (Geely EX2) sold 10,665 units in the first half of 2026.

6.2 Wad's "re-link with an MNC" option, realised

Wad's closing options for Proton were to design exportable brands for less demanding markets or to re-link with a global lead firm, "possible only by giving up local management control." He saw this as successful for Perodua (Daihatsu) and unsuccessful for Proton (Mitsubishi, which exited in 2004). The Geely deal is a second attempt at the re-link strategy, this time with a partner whose technology position in EVs is strong. It supports Wad's framework: the champion survives to the extent that a lead firm supplies the technology and platforms the champion cannot yet build. It is also a warning to the nationalist reading of his paper. The Proton that has revived is no longer an autonomous national technology project. It is a Malaysian marketing, assembly and distribution franchise anchored in a Chinese platform.

6.3 The protection caveat

The revival is also policy-assisted. Bloomberg reported on 6 May 2026 that Malaysia would restrict fully imported EVs from July by raising minimum retail prices and power-output thresholds, a move expected to push many mid-range Chinese imports out of the market. Wad documented a similar pattern: after the AFTA reforms Malaysia moved from tariffs to taxes that favoured the small and mid-sized segments where the national makers sat. Policy design continues to shape who wins in Malaysia. Chinese brands held over half of Malaysia's BEV market in May 2026 according to registration-based reporting, with Proton's e.MAS series alone about 35%, and BEVs reached 8.9% of the market in the first nine months. Proton is thus simultaneously a national champion and a Chinese-technology franchise.

7. Cross-examination III: Shanghai Stock Exchange lead firms in Southeast Asia

The Shanghai Stock Exchange (SSE) matters here because it lists two of the Chinese groups that have become Wad's new "lead geese": SAIC Motor (SSE: 600104) and Great Wall Motor (SSE: 601633, also listed in Hong Kong). I could not retrieve SSE exchange filings directly, so the evidence below comes from company disclosures as reported in the trade press and should be checked against the exchange filings before citation in formal work.

7.1 SAIC and MG in Thailand

SAIC reported 1.166 million overseas sales in the first nine months of 2026, up 52.4% year on year and already above its full-year 2025 total, with overseas volume about 37% of its 3.147 million total. Trade reports caution that these are sales counted by the company and are not equivalent to exports from China or local registrations. SAIC runs vehicle plants in Thailand, Indonesia and India. In Thailand, SAIC's venture with Charoen Pokphand (SAIC Motor-CP) and MG Sales (Thailand) reported 35,872 units of combined domestic sales and exports in 2025, up 44%. For 2026 MG targets 30,000 units and a 5% share. In June 2026 it launched a locally assembled MG URBAN electric hatchback at a starting price of THB 529,900.

The pattern is a close echo of Wad's logic: a lead firm with a captive pool of capital and technology localises assembly in response to local-content incentives and then uses Thailand as a hub.

7.2 Great Wall and the ex-GM plant

Great Wall acquired a General Motors plant in Rayong in 2020 and invested 22.6 billion baht to refit it, becoming the second Chinese maker to build in Thailand. At the time, trade reports said around 60% of the plant's output was sold in Thailand and 40% exported. Wad's Table 9 shows GM and Ford among the leading exporters from Thailand in the mid-2000s. Great Wall inherited an American multinational's industrial footprint. The ASEAN supply base built by American and Japanese lead firms is being re-used by Chinese ones, which is the clearest continuity between Wad's era and ours.

7.3 Implication for the flying-geese metaphor

Wad's critical perspective described Japanese assemblers and their suppliers as micro-level lead geese. A Chinese group listed in Shanghai and arriving with its own battery, parts and software ecosystem is a different kind of goose: a vertically integrated flock that may not need the Japanese tier-one suppliers (Denso, Aisin and similar) whose dominance Wad documented in Thailand, where he reports Japanese OEMs sourcing 70–90% from Japanese affiliates. If Chinese assemblers bring their own suppliers, Wad's finding that Thai suppliers were pushed into lower tiers may repeat itself, with a different set of foreign winners. This is a hypothesis, not a finding, and the supplier data to test it (tier counts, ownership, local-content shares by firm) are not yet public in comparable form.

8. The macro and credit overlay: Moody's

The Moody's Analytics economy.com research library was not directly accessible for this post, so I rely on Moody's published commentary as reported in trade press and on the topics of its Global Autos Market Update 2026.

Two points are useful. First, Moody's Ratings has maintained a negative outlook for global automotive manufacturing for the next 12–18 months, citing slowing growth, tariff uncertainty, supply-chain cost inflation and the higher variable costs of a rising BEV share, which it expects to reach slightly under 30% of global sales by 2030. It also forecasts Chinese domestic demand falling 4.6% as stimulus fades. A shrinking home market is a strong push factor for Chinese exports and overseas production, which helps explain the intensity of Chinese entry into ASEAN.

Second, Moody's earlier work flagged that auto and parts sectors in South and Southeast Asia are exposed to US tariff policy. That matters for Thailand, whose exports to the United States rose 44.3% in June 2026 per the FTI, and which faces a 19% reciprocal tariff plus sector-specific auto measures, according to press reports. Wad described ASEAN's attempt at regional integration and a free trade area with China; in 2026 the more binding constraint is a fragmented trade regime with a chain of overlapping tariffs and taxes. Credit conditions also matter at the household level: trade analysts attribute part of Thailand's earlier domestic slump to elevated household debt and car-loan rejection rates reaching about 70%, which kept consumer demand weak even as EV imports surged.

9. Synthesis: what survives, what breaks

Wad's claim (2009)	2026 evidence	Verdict
The 1997–98 crisis was the decisive turning point for Thai export orientation	Holds for 2000–07. A second turning point is visible in 2024–26 as Chinese BEV imports force localisation	Holds, but not unique
Thailand succeeds through the value chains of Japanese and American MNCs	Japanese share is under pressure. Chinese brands held 21.5% of the Thai market in 2025 (up from 4.0% in 2021), and Chinese EV brands were 88% of Thai EV sales	Partly outdated
It is "last call" for Proton: export niche or re-link with an MNC	Proton re-linked with Geely and its domestic share is at a multi-year high	Largely vindicated on mechanism, wrong on outcome
Industrial policy is not the decisive variable	Tiered excise taxes, local-content targets and EV import curbs are now central in both countries	Weakened
Local suppliers are marginalised in both countries	Plausibly continuing as Chinese assemblers localise, but comparable supplier data are missing	Open hypothesis
Wages and conditions track industry trajectories	No new comparable data gathered here	Not tested

The common thread is that Wad's actor-centred logic is robust: where a lead firm has the technology, sunk costs and a reason to export, Southeast Asian plants thrive. What is dated is his assumption about who the lead firms are, and about policy being secondary.

There is also a theoretical lesson. Wad's pessimistic prior (national champions are futile) was a prediction about autonomous national projects. The 2026 evidence suggests the relevant variable is the terms of dependence. Proton survived by accepting a foreign technology partner while keeping a domestic brand and market position, much as Perodua did with Daihatsu. That is neither full national control nor pure FDI-led assembly, and Wad's neat binary (FDI model versus national champion model) is too tidy for 2026. The two models are converging on a hybrid: local brand or local content requirements anchored in foreign technology.

10. Counterarguments and limitations

Is the Thai slowdown structural or cyclical? September 2026's 6.1% production decline coincided with floods, and import front-loading ahead of the 1 January tax distorts monthly sales and import data. A reasonable sceptic will say the Thai industry is simply digesting a shock and that a decade of data will look different. I take the view that the composition shift (Chinese lead firms, BEVs) is structural, while the 2026 volume path is partly cyclical.

Is Proton's recovery a policy artefact? A critic in Wad's tradition will point out that Proton's gains coincide with EV import curbs and tax preferences, and that the market share data come from different sources with different denominators (industry association volumes versus registrations). The accurate statement is that Proton's recovery is real but not independent of protection. Whether it can export at scale, which was Wad's actual test, is unproven in the evidence I gathered.

Is "success" the right metric? Wad treats export growth and net trade balance as proxies for viability. Those measures do not capture value-added, local ownership, supplier upgrading or worker outcomes, and he himself notes that high exports may mask low local value-added. A Thai plant exporting Chinese-designed BEVs may score well on units while capturing little design or battery value.

Source limitations. Wad's data are secondary and in places estimated, as he states. My 2026 data are mostly trade-press and company-reported figures. Bloomberg Intelligence and Moody's Analytics proprietary research were not accessible, SSE filings were not retrieved directly, and company "overseas sales" figures are not production or registration data. These findings should be treated as a structured hypothesis, not a replication.

11. Research agenda and conclusion

Four questions would sharpen the comparison:

1. Supplier ownership and tiering. Do Chinese assemblers in Thailand and Malaysia source from Chinese affiliates, Japanese keiretsu suppliers or local firms? Wad's Table 10-style firm counts, updated, would answer this.
2. Export performance of Proton. Wad's test was exports to emerging markets. Registration data in Malaysia say little about this.
3. Local value-added in BEVs. Battery and powertrain localisation determines whether Thai and Malaysian plants capture more than assembly margin.
4. Labour outcomes. Wad's wage and union analysis has no modern counterpart in this post and deserves its own study.

Conclusion. Seventeen years on, Wad's paper reads as a good map of a territory that has since moved. His mechanism, in which sunk-cost multinationals, local capability and domestic political economy jointly determine outcomes, explains both Thailand's tax-driven push to localise and Proton's Geely-powered revival. His specific predictions fare more mixed: Thailand's success is no longer a Japanese story, the policy-is-secondary finding looks dated, and Malaysia's "last call" turned into a second act. The most defensible update is that Southeast Asian auto industries are best understood as host platforms for whichever lead firms hold the next technology, and that the political task for host states is to bargain over the terms of that hosting: local content, supplier development, and eventually design and battery capability.

Sources

Anchor paper: Wad, P. (2009). The automobile industry of Southeast Asia: Malaysia and Thailand. Journal of the Asia Pacific Economy, 14(2), 172–193. DOI: 10.1080/13547860902786029.

Bloomberg and BloombergNEF (public pieces): "Malaysia Moves to Curb EV Imports in Blow to Chinese Automakers" (6 May 2026); "Thailand Moves to Tax Imported EVs More Than Local Models" (10 Sep 2026); "Chinese EV Imports Surge in Thailand Ahead of Planned Excise Tax" (28 Oct 2026); "Thailand EV Tax Gets Carmakers Weighing More Local Production" (6 Nov 2026); BloombergNEF, Electric Vehicle Outlook 2026.

Moody's: Moody's Ratings global automotive sector outlook (as reported by Autocar Professional); Moody's Ratings note on US tariff exposure in Asian sectors (as reported by BusinessToday, March 2025); Moody's Analytics, Global Autos Market Update 2026 (event description).

Shanghai-listed firms (company data via trade press): Gasgoo and EqualOcean reports on SAIC Motor overseas sales (Oct 2026); MarkLines reports on MG/SAIC Motor-CP in Thailand (2026); MarkLines report on SAIC and Great Wall foreign investments.

Thai and Malaysian industry data: Federation of Thai Industries production reports via MarkLines, Business Recorder, and just-auto (2026); Hong Leong Investment Bank research via Bernama and Business Today; paultan.org (July and October 2026); Gasgoo on Chinese brands in Thailand; Counterpoint Research; Viet Nam Investment Review on Thai EV supply-chain pledges.

All figures are as reported in the sources above and were retrieved on or before 8 November 2026. Verify against primary filings before formal citation.

 

The Intelligence Revolution: How Artificial Intelligence is Transforming Media

A Comprehensive Data Analysis of Artificial Intelligence Adoption in Media

 Based on: Chan-Olmsted, S.M.  (2019).  A Review of Artificial Intelligence Adoptions in the Media Industry.  International Journal on Media Management.  

 DOI: 10.1080/14241277.2019.1695619  

 Published: November 25, 2019

 ---

https://2024.sci-hub.box/7791/035a94fbb605ee66270bdd3b0daeb323/chan-olmsted2019.pdf

 

  Executive Summary

 The media industry stands at a critical inflection point.  Artificial intelligence has emerged not merely as another technological tool, but as a transformative force reshaping how content is created, distributed, and consumed.  This comprehensive analysis examines the adoption patterns, strategic implications, and challenges facing media companies as they integrate cognitive technologies into their operations.

 This study shows that AI applications in media have converged around eight primary functional areas, with adoption rates varying significantly across various media sectors and organizational types, based on systematic qualitative analysis of 149 industry reports, articles, and whitepapers from 2015 to 2018. ---

  1.  The Role of Artificial Intelligence in Media 

1.1 The Landscape of Competition The competitive dynamics have fundamentally changed as a result of the entry of technology giants like Netflix, Amazon, Apple, Google, and Facebook into the conventional media industry. Traditional media organizations must now match these platform companies' sophisticated AI capabilities or risk becoming obsolete. Key Market Setting: - Audience fragmentation is unprecedented for media companies. - Consumer expectations for personalized experiences have reached all-time highs

 - The volume and velocity of content production continue to accelerate

 - The business models for traditional media are constantly being disrupted. 

 1.2 The Importance of AI for Media Media presents unique difficulties, in contrast to the tangible goods industry, where AI applications are simpler: - Complexity of Content: Media products require judgment, interpretation, creativity, and nuanced communication

 - "Experience Goods Nature": Content value varies depending on the context. - "Social Impact Considerations": Media products have a significant impact on society. - Dual Product Nature: Media serves advertisers as well as audiences. ---

Media Industry Turning To Blockchain Technology And Artificial Intelligence

 2.  Research Methodology and Data Collection

  2.1 Systematic Qualitative Analysis Approach

 The research employed thematic analysis of industry materials, allowing for pattern identification across diverse AI applications.  This approach proved particularly valuable given the emerging and rapidly evolving nature of AI adoption in media.

 Parameters for Data Collection: - The period from 2015 to 2018 (the period of AI's commercial breakthrough) - Sample Size: 149 articles, industry reports, and whitepapers

 - Netflix, the New York Times, the Washington Post, the Associated Press, Apple, Hulu, YouTube, Facebook, Amazon, the BBC, the Guardian, Disney, and a variety of startups are among the organizations that have been studied. - Analysis Framework: Porter's Value Chain adapted for media industry

  2.2 Why 2015 as the Starting Point?

 A turning point in AI's commercial applications occurred in 2015: - Amazon launched Alexa

 - Machine learning applications significantly increased in size. - Deep learning neural networks achieved breakthrough performance

 - Industry-wide increases in AI technology investments ---

 3.  Core Findings: Eight Functional Areas of AI Application

 3.1 Audience Content Recommendations and Discovery

 Adoption Leader: Netflix

 Insights and Key Statistics: - This represents the most mature and widespread AI application in media

 - AI is used by 59% of news media worldwide to improve content recommendations. - Netflix's entire brand identity centers on personalized, algorithm-driven content delivery

 Technical Implementation:

 - Patterns of viewing are looked at by machine learning algorithms. - Natural language processing interprets content themes and user preferences

 - Content's visual elements are evaluated by computer vision. - The accuracy of recommendations is continuously improved by deep learning. 

**Business Impact:**

 - Increased user engagement and retention

 - Less effort required to discover content - Increased differentiation from rivals - Increased content ROI via improved matching 

**Example from Netflix:** 

Each subscriber's experience is algorithmically customized by the platform, including: - Homepage row selection and ordering

 - Specific titles featured in each row

 - Movies' visual thumbnails (customized for each user) - Recommendation logic for similar content

 Similar technologies address the historically poor discoverability of audio content by making contextual podcast searches and on-demand content discovery possible for audio platforms. 

3.2 Audience Engagement

 Adoption Rate Data:

 - 47% of media executives think AI will make their jobs easier. - 62% believe that AI will make it easier to make decisions. - Thirty percent say they don't know enough about AI applications. Three Primary Engagement Mechanisms:

 A.  Interactions in the context - Metadata are processed by machine learning to determine the best time to engage - Correlation between current events and user context in real time - Example: Washington Post's Knowledge Map connects ongoing stories with relevant background information seamlessly

 B.  Timely Engagement

 - AI makes it possible to respond in real time to audience preferences and behaviors - Automated alerts based on user interest profiles

 - Dynamic content prioritization based on trending topics

 C.  Assistants Powered by AI - Virtual assistants understand linguistic features and user intent

 - Voice-based interfaces enable natural content searching and navigation

 - Example: NBA teams deploy chatbots for game information, statistics, and ticket inquiries on social media

 Strategic Value:

 Contextual and timely engagement builds deeper brand relationships in today's fragmented media landscape. AI enables scale without sacrificing personalization—a critical competitive advantage.

 3.3 Augmented Audience Experience

 Technical Applications:

 - Enhanced image accuracy and processing

 - Improved comprehension of user intent and input - Improved context analysis and content correlation - Enhanced experiences in augmented and virtual reality Broad-Based Industry Adoption: Major platforms implementing experience augmentation:

 - Netflix, Hulu, Apple, Facebook, YouTube

 - Sports Illustrated, CBS All Access

 - iHeartMedia (audio platform)

 Specific Implementations:

 Optimization of the video:- AI-enhanced encoding improves streaming quality

 - Buffering is avoided with adaptive bitrate algorithms - Improvement of visual quality based on network conditions Enhancement of the audio: - iHeartMedia's Super Hi-Fi AI creates customized song transitions

 - Track-wide automated volume normalization - DJ-like experience at scalable rates for streaming

 User Interface Optimization:

 - Personalized interface layouts based on usage patterns

 - Contextual menu arrangements

 - Predictive content positioning

  3.4 Message Optimization

 Primary Focus: Advertising and marketing campaign personalization

 Key Benefits:

 - For Advertisers: Better audience matching, improved ROI, enhanced transparency

 - For Audiences: less ad fatigue and more relevant advertising - For Media Companies: Increased ad revenue, improved user satisfaction

 Technical Skills: - Real-time video indexing and analysis

 - Contextual ad placement within content

 - Modeling and tracking of attributions - Ad placement that changes based on user profiles Industry Impact - Digital Audio:

 AI makes it possible for digital audio advertising to compete with digital display advertising in the following ways: - Targeting precision

 - Campaign monitoring - Attribution modeling

 - Performance measurement

 This represents a significant competitive advancement for audio media platforms historically disadvantaged in digital advertising markets.

  3.5 Content Management

 Adoption Statistics:

 - 47% of media companies use AI for automated metadata creation

 - 77% of companies with large content libraries employ AI for content tagging

 - 36% use AI to measure and control quality. 

Primary Functions:

 A.  Extraction and Aggregation of Content - Automated collection from multiple sources

 - Intelligent parsing and categorization

 - Example: BBC's Juicer API aggregates news from BBC and external sources, automatically tagging content for searchability and trend analysis

 B.  Content Tagging and Indexing

 - Automated metadata generation

 - Simplified content searches

 - More quickly retrieving clips - Critical for organizations with extensive content archives

 C.  Control and Monitoring of Quality - Automated checking for rendering errors

 - Identifying errors in editing - Omission detection

 - Technical quality assurance

 D.  Content Classification and Sentiment Analysis

 - New York Times uses Perspective API for comment toxicity classification

 - Facebook employs machine learning for fake news identification

 - NewsWhip is used by the Associated Press to identify social media trends. 

E.  Content Adaptation

 - The production of clips automatically for social media platforms - Format optimization for different distribution channels

 - Viewer engagement enhancement through platform-specific content

 Business Value:

 Content management represents one of the highest-value AI applications for media companies with extensive libraries.  The efficiency gains enable human resources to focus on higher-value creative and strategic activities.

 3.6 Creation of Content Revolutionary Potential: AI augments and automates various content creation activities

 Applicable Creative Processes:

A.  Post-Production Automation

 - Plot identification and scene selection

 - Automated scripting assistance

 - Example: IBM Watson produced a cognitive movie trailer for "Morgan" (2016)

 - Hollywood studios use AI to make trailers and translate content. 

B.  Robotic Journalism- Automated earnings reports and coverage of minor league baseball from the Associated Press - Washington Post's Heliograf: Expanded election coverage with predetermined parameters and geo-targeting

 - Yahoo Sports: Automatic Insights platform generates narratives from sports data

 C.  News Production Efficiency

 - Writers focus on high-profile content while AI covers comprehensive baseline reporting

 - Data-centric stories generated automatically from structured datasets

 - Real-time data visualization (Reuters' Graphiq semantic technology)

 D.  Tools for Content Optimization - Forbes' AI suggests topics for articles based on previous work by authors. - The creation of headlines using image sentiment analysis - Rough draft creation for contributors to refine

 E.  Sound Design and Audio Production

 - Disney Research: Automated association of sound effects (such as car images and engine sounds) - Applications for targeted audio experiences for specific audience segments

 The Implications for Strategy: AI makes it possible to divide up work so that: - Humans focus on creative storytelling, intricate analysis, and profound insights. - Machines deliver comprehensive, timely, data-driven content at scale

 - Media organizations expand coverage without proportional resource increases

  3.7 Audience Insights

 Strategic Focus: Internal operations and marketing strategy optimization

 Primary Applications:

 A.  Marketing Analytics

 - Enhanced attribution tracking and modeling

 - Programmatic advertising optimization

 - Marketing forecasting with audience data integration

 B.  Market Segmentation- The creation of personas powered by AI using huge datasets like:  - Geo-specific events and behaviors

  - The patterns of on-site interaction  - Referral source analysis

  - Psychographic factors

  - Past purchasing habits  - Communication preferences

 C.  Advanced Analytics - Forecasting content performance - Audience growth forecasting

 - Revenue optimization modeling

 D.  Real-Time Emotion Tracking

 - Example: Disney's facial scanning software tracks audience emotions during test screenings

 - Deep learning analyzes thousands of subjects simultaneously

 - Helps make decisions about content creation and editing 

E.  Pay-Per-Click Optimization

 - Discovery of new advertising channels

 - Platform testing and optimization

 - Advanced targeting refinement

 Business Impact:

 AI-powered insights enable precision marketing at scale, reducing waste and improving ROI across all marketing investments.  Content creation and distribution strategies are transformed when it is possible to pinpoint the preferences of an audience. 

3.8 Automation of Operations Foundation: Utilizes efficiency enhancement to support numerous other AI applications. Automation of mundane, repetitive tasks to free up human resources for complex, strategic work is the main benefit. Primary Implementations:

 A.  Automation of Content Production - Automated captioning, which is widely used in the industry. - Event monitoring in real time - Social media management

 - As an illustration, AI reduced the Associated Press's weekly copy editing time by 800 hours. **B.  Control of the content - New York Times: AI moderates more article comments faster than human moderators alone

 - Policy violations and harmful content are discovered by machine learning - Human moderators focus on nuanced, complex cases

 C.  News Verification and Research

 - Reuters' News Tracer: Tracks breaking news on Twitter automatically

  - Analyzes percentage of daily tweets for breaking news

  - Utilizes profile and network analysis to verify the authenticity of information  - Reverse-engineers the procedures for journalistic verification 

D.  Fact-Checking Systems

 - Voyc: Voice-scanning AI identifies questionable statements in real-time

  - Converts live audio into text  - Comparisons of verified fact databases with cross-references  - Sends alerts for statements conflicting with verified information

 - The Full Fact and the First Draft: ML-based automation to combat misinformation 

E.  Rights Management

 - Content fingerprinting for identification

 - Micro-licensing that incorporates blockchain - Usage tracking and payment processing

 - Viewing behavior validation

 - Verification of the event's actuality - Unauthorized distribution protection

 F.  Media Management - Media monitoring automation

 - Transcription services

 - Ad verification systems

 - Reference clip generation

 Efficiency Impact:

 The most immediate and measurable ROI for AI investments is operational automation. Media companies with intensive content operations report significant cost reductions and productivity improvements.

 ---

 4.  Trends in Adoption Across Media Industries 

4.1 Sector-Specific Adoption Rates

 Results of the Survey: 

By Type of Media - Cable Systems and MSOs: 44% (highest) adoption - Cable Networks: 30% adoption

 - Broadcast Networks: 25% adoption

 - Organizations that publish news online: high adoption (the exact percentage is unknown, but it is among the most active) - Audio/Radio Platforms: On the rise By Application Type (Global Survey of News Media): - Content recommendations: 59%

 - Workflow automation: 39%

 - Optimization for commercial purposes (ad targeting, dynamic pricing): 39% - Story support provided by intelligent agents: 35% 4.2 Pioneers in the Use of AI Organizations that are most active: - Netflix (content recommendations, personalization, production decisions)

 - New York Times (content moderation, tagging, annotation)

 - The Washington Post (automated reporting and testing of headlines) - Associated Press (automated journalism, content management)

 - BBC (management of metadata and content aggregation) **Common Characteristics of Leaders:**

 - Operations that are digital-native or digital-first - Large-scale content production operations

 - Direct-to-consumer distribution models

 - Significant technical infrastructure investment

 - Data-driven organizational culture

 4.3 Barriers to Adoption

 Primary Obstacles Recognized: Integration Difficulties:

 - 47% of respondents cite difficulty integrating AI with existing processes/systems

 - Problems with legacy systems' compatibility - Organizational silos preventing data integration

 Constrained Resources: - Infrastructure costs are a problem for smaller businesses. - A lack of technical knowledge - Insufficient data volume for effective AI training

 Knowledge Gaps:

 - One-third of media practitioners lack confidence in understanding AI applications

 - Organizational insufficiency in algorithmic literacy - Training and development gaps

 Cultural Concerns:

 - 47% concerned about reduced human control

 - 55% worried about trustworthy insights

 - Conflict between technical and creative teams - Resistance to algorithm-driven decisions

 ---

 5.  Analyses of the Strategic Value Chain  

5.1 Porter's Value Chain Applied to Media

 The research adapts Porter's generic value chain framework to analyze AI's strategic value in media.  This framework divides business activities into primary and support functions, enabling systematic evaluation of AI's impact.

 Modified Media Value Chain Components:

 Primary Activities:

 1.  Acquisition of raw materials (content, data, and media assets) for inbound logistics 2.  Operations: Content transformation and production

 3.  Outbound Logistics: Content delivery and distribution 4.  Marketing/Sales: Audience acquisition and engagement

 5.  Service: relationship management and post-purchase support Promotional Activities: 1.  Acquisition of input and resources through procurement 2.  Management of Human Resources: Recruiting and Developing People 3.  Technology Development: Systems and knowledge management

 4.  Infrastructure: General management and coordination functions

 Critical Modification: Integration of audience feedback loop recognizing that audiences create value through user-generated content and audience insights for advertisers.

 5.2 AI Value Creation Across Primary Activities

 Inbound Logistics Enhancement:

 - Quick and efficient identification and storage of raw materials (such as data, images, and videos) - Content licensing and sourcing that is automated - Systems for intelligent asset management - Rights tracking and verification

 Operations Transformation (Highest Value):

 - Content creation augmentation

 - Improving production workflow - Quality assurance automation

 - Resource allocation optimization

 - Performance: Better perception and productivity Innovation in Outbound Logistics (Highest Value): - Personalized content distribution

 - Optimization for multiple platforms - Enhanced delivery experiences

 - Real-time distribution adjustments

 - Performance: Enhanced effectiveness and innovation

 Marketing/Sales Excellence (Highest Value):

 - Scalable insights into audiences - Individualized approaches to engagement - Predictive campaign optimization

 - Real-time message adjustment

 - Performance: Improved cognition and effectiveness

 Service Transformation:

 - Timely audience interactions (previously impossible in one-way media systems)

 - Chatbot customer service

 - As a service, customized recommendations - Problem solving before it happens  5.3 AI Value Creation in Support Activities

 Technology Development (Primary Support Focus):

 - Improved management of information and knowledge - Improved data processing capabilities

 - A sophisticated infrastructure for analytics - Scalable technology platforms

 "Infrastructure Improvement:" - Data integration across departments

 - Decision-supporting capabilities from analytics - Coordination systems improvement

 Limited Impact Areas:

 - Procurement procedures (currently, a minimal AI application) - Human resource management (limited AI deployment in talent functions)

 5.4 Media Workflow Transformation

 AI impacts all stages of the adapted media workflow:

 Pre-Production:

 - Improving workflow planning - Forecasting the allocation of resources - The development and validation of content concepts

Production:

 - Facilitation of collaboration - Real-time quality monitoring

 - Automated technical processes

 Post-Production:

 - Rich content indexing

 - Metadata analytics

 - Help with editing 

Media Lifecycle Management:

 - Automating asset management - Rights monitoring - Archive optimization

 Multi-Platform Distribution:

 - Distribution orchestration

 - Format optimization

 - Delivery personalization

 Experience Integration:

 - An individual presentation - Interaction measurement

 - The integration of feedback 

Consumption:

 - Real-time optimization

 - Experience enhancement

 - Performance tracking

 ---

 6.  Strategic Implications and Competitive Dynamics

  6.1 The End of Economies of Scale

 Paradigm Shift:

 The advantages of traditional media that are based on scale are decreasing because: - Technology platforms rented out (such as Amazon AWS) - AI democratization through APIs and services

 - Reduce obstacles to the creation and distribution of content - Reduced fixed costs for market entry

 New Competitive Parameters:

 - Agility over size

 - Innovation over infrastructure

 - Niche expertise over mass reach

 - The power to personalize over broadcast power 

Strategic Implications for Incumbents:

 - Must shift from scale-based to innovation-based strategies

 - Need to develop new bundling and aggregation approaches

 - Focus on product excellence over distribution dominance

 - Embrace "unscaled" nimble competitor strategies

 6.2 Value Creation Through AI

 Three Primary Value Mechanisms:

 A.  Perception Enhancement

 - Image and speech recognition improvements

 - Content analysis and categorization

 - Quality assessment automation

 - Pattern identification in large datasets

B.  Cognition Augmentation

 - Improved decision-making speed and accuracy

 - Complex data interpretation and analysis - Predictive modeling capabilities

 - Automated judgment on routine matters

 C.  Prediction Optimization

 - Audience behavior forecasting

 - Content performance prediction

 - Resource allocation optimization

 - Risk assessment and mitigation

 6.3 Differentiation: Content Creators vs.  Distributors

 Content Creators (Production Companies, News Organizations):

 Primary Value Areas:

 - Operational efficiency (improved perception)

 - Content decision enhancement (improved cognition) - Workflow automation

 - Quality improvement

 Important Applications: - Content creation assistance

 - Production optimization

 - Audience insights for content development

 - Archival administration Content Distributors (Platforms, Networks, Streaming Services):

 Primary Value Areas:

 - The efficiency of marketing (improved cognition) - Service innovation (enhanced prediction)

 - Audience relationship building

 - Enhancement of distribution Important Applications: - Recommendations for content - Engines that customize - Attraction of the audience - Enhancement of messaging Vertically Integrated Media Companies (Hybrid Organizations): - Must excel across all value areas

 - Face greatest integration challenges

 - Require an even distribution of expenditures across applications - The need to control conflicts between distribution and production priorities ---

 7.  Important Obstacles to AI Implementation 

7.1 Finding a balance between human and machine intelligence The Netflix Case Study:

 The conflict over "Grace and Frankie's" marketing strategy exemplifies the tension: - Tech team advocated for algorithm-optimized promotional images (highest response ROI)

 - Creative team concerned about star alienation and contract violations

 - Instead of Jane Fonda as the main character, images featured supporting characters. Broader Implications:

 - Data-driven decisions versus creative choices - Building relationships versus measuring efficiency - Brand value over time versus performance in the short term - Soft relational metrics vs. hard performance data

 "Survey Findings:" - 47% of media executives concerned about reduced human control

 - 55% worried about trustworthiness of AI-generated insights

 - Minority believe AI will negatively impact job availability (exact percentage not quantified)

 Strategies for Getting Things Done: - Establish clear decision-making frameworks

 - Define domains for AI and human judgment to compete in. - Create hybrid decision processes

 - Develop AI literacy across organizations

 - Build cultural bridges between technical and creative teams

 7.2 Efficiency vs.  Effectiveness Trade-offs

 Customer Interaction Challenges:

 Efficiency Gains:

 - Chatbots handle high volumes of routine inquiries

 - Virtual assistants provide 24/7 availability

 - Automated responses ensure consistency

 - Cost per interaction dramatically reduced

 Effectiveness Limitations:

 - There is currently no emotional intelligence in AI. - Unable to show genuine empathy - Limited comprehension of the context - May damage brand relationships if overused

 Privacy and Ethics Concerns:

 - Personal data collection necessary for personalization

 - Unclear ethical boundaries for data usage

 - Regulatory uncertainty (especially pre-GDPR implementation in study period)

 - Consumer trust implications

 Problem with Matching Content: - Content is efficiently matched to preferences by algorithms. - On the other hand, preferences may not coincide with societal good - Echo chamber and filter bubble concerns

 - Responsibility for content exposure decisions

 Questions of a Strategic Nature: - How much automation is appropriate in audience relationships?

 - When should human interaction be preserved?

 - How to differentiate assistant interactions from friend interactions?

 - What ethical guidelines should govern AI use?

 7.3 Data, Integration, and Competency Requirements

 Problems with Access to Data: Volume Requirements:

 - For AI to work, huge datasets are needed. - Smaller media organizations at disadvantage

 - Data acquisition is expensive. - Historical data may not be available

 Effects on Privacy: - Tension between data needs and privacy protection

 - Regulatory compliance requirements

 - Consumer consent management

 - Data security obligations

 Integration Barriers:

 Compatibility of the System: - 47% of organizations struggle with AI integration into existing processes/systems

 - Legacy system limitations

 - Siloed data across departments

 - Technical debt from historical IT investments

 Cost Considerations:

 - Needs for infrastructure investment - Ongoing maintenance expenses

 - Small organizations particularly challenged

 - Uncertainty regarding ROI in the early stages of implementation Competency Gaps:

 Algorithmic Literacy:

 - Insufficient understanding across organizations

 - Insufficient resources for training - Rapid technology evolution outpaces learning

 - Need for transparency in AI operations

 Management Capabilities:

 - Top challenge: lack of business alignment

 - Insufficient change management skills

 - Lack of familiarity with AI-enhanced workflows - Need for new leadership competencies

 Technical Expertise:

 - A shortage of AI professionals who are qualified - Competition from technology companies for talent - Timelines for existing employees' training - Retention challenges

 7.4 The evolutionary versus Revolutionary Expectations

 Reality Check:

 Current AI Limitations:

 - Deals with defined tasks, not complex judgment

 - Cannot replicate human synthetic reasoning

 - Lacks contextual understanding

 - Limited to narrow applications

 Market Fragmentation:

 - Many vendors offering point solutions

 - The absence of extensive platforms - Integration challenges across systems

 - The difficulty of selecting vendors Both objectivity and bias: - Biases in training data and parameters are reflected in AI. - The importance of human objectivity in the design phase - Discrimination by accident poses a risk - Requirement for ongoing monitoring and adjusting 

Strategic Recommendations:

 - Maintain realistic expectations about AI capabilities

 - Prioritize augmentation over replacement. - Implement applications that enable humans to work more effectively

 - Plan for evolutionary adoption rather than revolutionary transformation

 - Build organizational learning and adaptation capabilities

 ---

 8.  Industry Survey Data Analysis

 8.1 European Media Practitioner Survey

 Overall AI Impact Perception:

 - Eighty percent agree that AI will significantly affect the media industry. - 62% believe AI will improve decision-making

 - 47% anticipate increases in productivity. - One-third lack confidence in understanding AI applications

 Primary Concerns:

 - Less human oversight: 47% - Trustworthy insights: 55%

 - Job availability impact: Minimal concern (specific percentage not provided)

 Interpretation:

 High awareness and optimism coexist with significant knowledge gaps and control concerns.  This suggests need for education and change management focus in AI implementation strategies.

 8.2 Cognitive Technology Adoption Survey (Deloitte)

 Manager Enthusiasm:

 - 87% consider cognitive technologies important to product offerings

 - 92% view AI as vital to internal business processes

 - 76% believe cognitive technologies will transform their businesses

 Future Outlook:

 Particularly high expectations in B2C industries, where direct consumer interaction makes AI applications more immediately valuable.

 Top Implementation Challenges:

 1.  Lack of business alignment is the primary obstacle. 2.  Integration difficulty with existing processes/systems

 3.  Cost considerations

 4.  Manager competency gaps

  8.3 Use of AI in Specific Media Content Management Applications:

 - 47% use AI to create metadata automatically. - 77% with large content libraries employ AI for tagging

 - 36% use AI for quality control and measurement

Automated Captioning:

 - Widespread adoption across industry

 - Considered essential rather than optional technology

 - High satisfaction with current capabilities

 Social Media and Live Events: - Increasingly important application area

 - The value of real-time monitoring capabilities - It is essential to integrate with social media platforms. 8.4 Most Prominent AI Applications (Technology Research, 2018) 

Top Applications Across Industries:

 1.  Security-related applications

 2.  Analytics and business intelligence

 3.  Customer service/relationship management

 Media-Specific Priority Applications:

 - Content recommendations (highest adoption)

 - Process automation - Commercial optimization

 - Intelligent content discovery assistance

 8.5 Adoption Comparison: Media Sectors

 By Organization Type:

 - Cable Systems/MSOs: 44% adoption

 - 30 percent adoption of cable networks - Television Networks: 25 percent adoption 

Analysis:

 Distribution-focused organizations show higher adoption rates than pure content creators.  This aligns with the finding that AI creates more immediate value in distribution and audience engagement functions.

 By Content Library Size:

 Organizations with large content archives show dramatically higher AI adoption (77% for metadata management) compared to those with smaller libraries, suggesting economies of scale in AI implementation.

 ---

 9.  Future Implications and Research Directions

  9.1 Emerging Trends Beyond 2018

 Technology Evolution:

 - Neural network advancement enabling deeper contextual analysis

 - Improved intent recognition in natural language processing

 - Enhanced computer vision for richer content analysis

 - More sophisticated emotion and sentiment detection

 Application Expansion:

 - From basic automation to complex decision support

 - From siloed applications to integrated AI systems

 - Capabilities that shift from reactive to predictive - From efficiency focus to innovation enablement

  9.2 Research Opportunities

 Studies of Measurement and Impact: - Quantifying AI ROI in media contexts

 - Using AI-enhanced experiences to gauge audience satisfaction - The effects of tracking creative quality - Evaluating the duration of competitive advantage Progression of the Concept: - Media industry-specific AI adoption models - AI integration in the evolution of the value chain - Human-machine collaboration frameworks

 - The theory of decision-making in AI-augmented environments Studies on Implementation: - AI integration best practices - Change management approaches

 - Organizational structure adaptations

 - Skill development pathways

 Ethical and Legal Research:

 - Privacy protection frameworks

 - Systems of algorithmic accountability - Avoidance and detection of bias - Regulatory compliance models

 Cultural and Social Impact:

 - The impact that media mediated by AI has on democratic discourse - Cultural implications of algorithmic content curation

 - Effect on journalism standards and quality - Changes in the information diets of the audience 

 9.3 Media Organizations' Strategic Recommendations For Content Creators:

 1.  Automating workflows should be prioritized to free up creative talent for higher-value work. 2.  Invest in content management AI for organizations with significant archives

 3.  Develop AI-augmented creation capabilities while maintaining human creative control

 4.  Build data infrastructure to support content performance analysis

 For Distributors of Content: 

1.  Concentrate on personalization engines as the primary competitive advantage. 

2.  Enhance audience insights capabilities for better content matching

 3.  Develop sophisticated engagement strategies using AI-powered tools

 4.  Invest in experience optimization across all platforms and touchpoints

 For Hybrid Businesses:

 1.  Establish clear governance for AI versus human decision-making. 

2.  Build bridges between creative and technical teams

 3.  Develop integrated data strategies across production and distribution

 4.  Balance efficiency and effectiveness based on strategic priorities

 All-inclusive Recommendations: 

1.  Start with high-impact, lower-risk applications to build capability and confidence

 2.  Invest in algorithmic literacy and organizational learning

 3.  Develop ethical frameworks for AI use before implementation pressures arise

 4.  Instead of radical transformation, "plan for evolutionary adoption." 

5.  Establish connections with technology providers and research organizations. 

6.  Monitor emerging technologies and competitive developments continuously

 ---

 10.  Conclusion: Navigating the AI-Transformed Media Landscape

  10.1 Important Takeaways 

Ai as a Technology for All Purposes: Like the internet before it, AI represents a transformational technology applicable across all media functions and sectors.  The question for media organizations is not whether or not to implement AI, but rather how to do so in an ethical and strategic manner. Eight Core Application Areas:

 The research identifies eight primary domains where AI creates value in media:

 1.  Content recommendations and discovery

 2.  Audience engagement

 3.  Experiences enhanced 

4.  Message enhancement 

5.  Management of content 

6.  Content creation

 7.  Audience insights

 8.  Operational automation

 Three Value Creation Mechanisms:

 AI delivers value through:

 1.  Automation: reducing labor-intensive, repetitive tasks 2.  Insights: Using data to generate useful information 3.  Engagement: Facilitating individualized and scalable interactions Primary vs.  Activities to Aid: AI creates the most value in primary activities—particularly operations, outbound logistics, and marketing/sales—while technology development leads support activity transformation.

 10.2 The Act of Balancing Success in AI adoption requires media organizations to navigate multiple tensions:

 Creative vs.  Algorithmic Decision-Making:

 Finding the right balance between data-driven efficiency and relationship-focused effectiveness remains the central challenge.  Organizations must develop frameworks that leverage AI strengths while preserving human judgment where it matters most.

 Efficiency in the Short Term versus Long-Term Effectiveness:

 While AI delivers immediate operational benefits, media companies must ensure these gains don't compromise brand relationships, content quality, or societal responsibilities.

 Scale Advantages vs.  Modest Innovation: Traditional scale-based competitive advantages are diminishing as AI and rented platforms democratize capabilities.  Media organizations must develop new sources of differentiation based on innovation, personalization, and niche expertise.

  10.3 Implementation Imperatives

 Build a solid foundation of competence: Organizations must invest in algorithmic literacy, technical infrastructure, and change management capabilities before expecting AI to deliver transformative value.

 Start strategically and think systemic: Rather than pursuing disconnected point solutions, media companies should develop comprehensive AI strategies aligned with business objectives and organizational capabilities.

 Make integration a top priority: The primary challenge isn't AI capability itself but integrating AI into existing processes, systems, and organizational cultures.  Success requires addressing technical, operational, and cultural dimensions simultaneously.

 Keep an ethical eye on things: As AI becomes more powerful and pervasive, media organizations bear responsibility for transparent, equitable implementation that serves audience and societal interests alongside business objectives.

  10.4 The Path Forward

 The media industry stands at a pivotal moment.  AI adoption is accelerating, competitive dynamics are shifting, and audience expectations continue evolving.  Companies that are able to successfully navigate this change will: - View AI as a business mindset and competency, not just a technology

 - Balance machine efficiency with human creativity and judgment

 - Invest in organizational capabilities alongside technical infrastructure

 - Maintain focus on audience value creation amid operational optimization

 - Develop ethical frameworks that build trust while enabling innovation

 The evidence presented in this analysis demonstrates that AI is no longer a future possibility but a present reality reshaping media industry competitive dynamics.  The winners in this transformed landscape will be organizations that approach AI adoption strategically, implement it thoughtfully, and leverage it to enhance rather than replace human creativity, judgment, and relationship-building capabilities.

  10.5 Final Reflection

 As artificial intelligence continues its rapid evolution, the media industry faces both tremendous opportunities and significant challenges.  According to the findings of this analysis, the use of AI in media has progressed from its experimental stages into a practical reality in numerous functional areas. The data shows clear patterns: leading organizations prioritize AI in audience-facing applications while using it to enhance operational efficiency behind the scenes.

 However, the journey toward AI-transformed media operations is evolutionary, not revolutionary.  Success requires realistic expectations, sustained investment, organizational adaptation, and most importantly, thoughtful integration that preserves the creative, relational, and societal value that distinguishes media from other industries.

 The question is no longer whether AI will transform media, but how organizations will shape that transformation to serve audiences, support creative excellence, and fulfill media's crucial role in society.  Those who approach this challenge strategically, ethically, and with appropriate balance between human and artificial intelligence will be best positioned to thrive in the intelligence revolution reshaping the media landscape.

 ---

  References

 Chan-Olmsted, S.M.  (2019).  A Review of Artificial Intelligence Adoptions in the Media Industry.  International Journal on Media Management, 21(3-4), 193-215. https://doi.org/10.1080/14241277.2019.1695619

 ---

 This analysis is based on research published in the International Journal on Media Management (ISSN: 1424-1277 Print, 1424-1250 Online).  The original research employed systematic qualitative analysis of 149 industry sources from 2015-2018 to identify patterns in AI adoption across media organizations.