Ajinomoto ABF Film Price Hike of 30% Hits AI Chip Supply Chain; Over 50 Chinese Consumer Companies Enter Hard

In 1908, Professor Kikunae Ikeda of Tokyo Imperial University isolated glutamic acid from seaweed soup, and businessman Saburosuke Suzuki named it Ajinomoto and brought it to market. Through monosodium glutamate production, Ajinomoto accumulated industrial control capabilities in fermentation, separation, and purification, which were later applied to amino acid production, expanding its business from seasonings to food, pharmaceuticals, feed, and fine chemicals. Research teams discovered that byproducts of MSG production could be made into a resin-like synthetic material with high insulation, high durability, low thermal expansion, and easy processability, but the finding was shelved due to a lack of application scenarios at the time. In 1996, Intel sought to use Ajinomoto's amino acid technology to develop thin-film insulating materials for CPUs. Ajinomoto's team completed ABF prototypes and samples within four months, and after three years of development, Intel officially adopted the material. In the third quarter of 2026, Ajinomoto announced a 30% price increase for ABF, putting pressure on the global AI chip supply chain. Since the second half of 2025, multiple Chinese listed consumer companies have begun investing in hard technology. By-health's revenue fell from approximately RMB 9.4 billion in 2023 to about RMB 6.3 billion in 2025. Yangyuan Beverage's net profit in 2025 was RMB 1.26 billion, halved from its peak in 2018. Andeli faces pressure from sugar reduction trends, with the expansion of sugar-free tea drinks squeezing the concentrated juice market. Jinzi Ham stated in an investment announcement that its core business is developing slowly. ChatGPT in 2023 triggered a global AI boom, and DeepSeek's surge in 2025 further fueled the trend. Toread, with a dual main business of outdoor products and chips, saw its market value multiply more than six times in four years. Allbirds, renamed NewBird, pivoted to computing power leasing and surged 582% in a single day. Lotus Health, which began diversifying into computing power in 2023, saw its stock price rise from about RMB 6 to nearly RMB 13, with market value increasing by approximately RMB 12.5 billion. From 2023 to 2024, the three memory giants Samsung, SK Hynix, and Micron together recorded losses exceeding USD 50 billion. Consumer companies diversifying into hard tech generally follow three paths. The first is small financial equity stakes. In June 2026, By-health invested RMB 50 million in Yuanli Semiconductor, acquiring a 0.97% stake. Yuanli Semiconductor reported full-year 2025 revenue of RMB 3.7966 million and a net loss of RMB 62.9594 million; in the first quarter of 2026, it had no revenue and its quarterly loss expanded to RMB 10.9152 million, with a post-investment valuation exceeding RMB 5.1 billion. By-health stated in the announcement that the investment is purely financial, aiming to establish a cognitive window into frontier technology. In the first half of 2026, By-health made five investments in the tech sector, spending over RMB 300 million in total, with individual company stakes generally below 1%. The company had over RMB 2.3 billion in cash and cash equivalents, with an asset-liability ratio of about 20%. The second path is controlling acquisitions or industrial fund investments. Since 2021, Wahaha has invested through Wahaha Venture Capital in intelligent manufacturing, robotics, new materials, and biomedicine, and has invested in chip unicorn Muxi Integration. Tsingtao Brewery led the establishment of a RMB 1 billion industrial fund with a RMB 400 million contribution, focusing on cultural technology and consumer industry transformation and upgrading. Zhong Shanshan, through Yangshengtang, invested RMB 500 million for a 10% stake in solid-state battery electrolytes, and placed bets on embodied intelligence and photonic chips. SDIC Zhonglu invested RMB 6 billion to acquire all equity of China Electronics Engineering Design Institute, focusing on the design and construction of semiconductor fabs and computing centers. The third path is using AI and intelligent manufacturing to transform traditional industries. Haitian Flavouring was recognized by the World Economic Forum as the world's first lighthouse factory in soy sauce brewing. It independently developed AI bean face recognition to screen soybeans, and an AI electronic nose that collects over 170 aroma data points to build a soy sauce aroma library, moving away from traditional processes in the koji-making stage. Bosideng developed its own AI aesthetic brain to empower design and R&D, embedding AI across the entire chain. Anta plans to invest another RMB 20 billion over the next five years, focusing on professional sports, high-end outdoor, and winter sports technology. In the first quarter of 2026, Toread's chip business revenue reached RMB 179 million, up 206.92% year-on-year, accounting for 36.09% of total revenue. Toread's chip layout has strategic synergy with its outdoor main business, with self-developed chips used in terminal products such as smart exoskeletons, and its acquisition targets include display driver ICs, touch ICs, and other niche segments; after acquisition, it dispatched teams to integrate operations. Haitian Flavouring built a vertical large model, converting master craftsmen's experience into data models. Gree Electric Appliances built its own silicon carbide chip factory, achieving the use of self-developed chips in 1 million air conditioners with a yield rate above 99%. According to McKinsey reports from 2018 to 2025, the failure rate of traditional companies diversifying into hard tech ranges from 70% to 75%. Lianchu Securities' 2024 A-share M&A report shows that the failure rate of cross-industry high-tech M&A is 57.2%, about 2.5 times that of intra-industry consolidation M&A. Yuanli Semiconductor, in which By-health invested, had revenue of only RMB 3.7966 million in 2025 and a net loss of RMB 62.9594 million. Labixiaoxin Food's HKD 188 million acquisition of an AI company recorded losses of RMB 41.45 million in 2024 and RMB 20.88 million in 2025. Annil's cross-industry bet on AI computing power led to seven consecutive daily limit-ups, followed by a bubble burst, and its actual controller reduced its stake by over 59%.