Core Viewpoint - The article discusses the rise of AI-related side jobs, highlighting the disparity between the perceived ease of earning money through AI and the actual challenges faced by individuals in these roles. It emphasizes that many success stories are exaggerated and often serve the interests of training institutions rather than reflecting genuine opportunities for wealth creation [5][15]. Summary by Sections AI Side Jobs Overview - AI side jobs are categorized into three main types: metaphysical (e.g., AI fortune-telling), entertainment and consumer goods (e.g., AI stickers and avatars), and practical applications (e.g., AI writing and video production) [7][9][13]. - The demand for AI writing and video production is growing, but the market is becoming increasingly competitive and professionalized, limiting opportunities for non-professionals [16][17]. Challenges in AI Side Jobs - The demand for AI-related services is often overstated, as large companies are developing their own AI tools, reducing the space for casual participants [16]. - The low technical barriers to entry for AI side jobs lead to a saturation of the market, resulting in a decline in quality and increased competition [14][17]. The Role of Training Institutions - Many training institutions exploit the hype around AI to sell courses, often targeting vulnerable demographics such as young people and the elderly with exaggerated claims of potential earnings [18][21]. - The marketing strategies of these institutions often play on fears of missing out (FOMO) and the desire for quick financial gains, leading to impulsive purchases of training programs [21]. Future Outlook - As the AI market matures, it is expected to move towards greater standardization and professionalism, which may raise the entry barriers for side jobs [21]. - Individuals who genuinely master AI technologies will be better positioned to benefit from the evolving landscape, while those relying on superficial skills may struggle to compete [21].

Core Insights - The article discusses the rise of AI-related side jobs, highlighting the allure of easy earnings but contrasting it with the reality of challenges and competition in the market [1][12][18] Group 1: Types of AI Side Jobs - AI side jobs can be categorized into three main types: metaphysical (e.g., AI fortune-telling), entertainment/consumer (e.g., AI stickers and avatars), and practical (e.g., AI writing and video production) [2][5][9] - AI fortune-telling, particularly tarot card reading, has gained popularity as it allows inexperienced users to engage with AI tools to perform readings, creating a new market for novice practitioners [3][5] - Entertainment-related AI jobs, such as custom stickers and avatars, require minimal investment and can be easily produced using AI tools, with prices ranging from 9.9 to over 99 yuan depending on complexity [5][7] Group 2: Market Dynamics and Challenges - The demand for AI writing, video production, and marketing solutions is growing, but the market is becoming increasingly professionalized, limiting opportunities for non-professionals [13][14] - The low technical barrier for entry in many AI side jobs leads to a saturated market with high competition, resulting in a race to the bottom in terms of quality and pricing [11][14] - Many AI side jobs are portrayed as lucrative opportunities, but the reality often falls short, with most individuals earning only modest amounts, similar to traditional freelance work [11][12] Group 3: The Role of Training Institutions - The article highlights the emergence of training institutions that capitalize on the hype surrounding AI side jobs, often exaggerating potential earnings to attract customers [15][18] - These institutions target various demographics, using tailored marketing strategies to appeal to both younger and older audiences, often emphasizing the urgency to learn AI skills [15][16] - The proliferation of such training programs raises concerns about the quality and credibility of the information being disseminated, as many are primarily focused on selling courses rather than providing genuine value [18]

Core Insights - The article discusses the challenges of multi-turn conversations in AI applications, emphasizing that current models struggle with maintaining context and coherence over extended dialogues [3][5][6] - It highlights the necessity of designing Standard Operating Procedures (SOPs) to improve multi-turn dialogue experiences, focusing on goal-oriented interactions [10][12][48] Group 1: Challenges in Multi-Turn Conversations - Multi-turn dialogue is complex due to the tendency of users to open new windows for new questions, reflecting a lack of trust in the model's capabilities [4] - The forgetting rate of key information in AI models increases significantly with the number of dialogue turns, with a 37% information loss after 7 turns and 68% after 12 turns [5][6] - The use of pronouns in conversations complicates AI interactions, as models may misassociate these terms, requiring users to provide additional context [7] Group 2: Importance of SOP Design - Effective multi-turn dialogue requires clear goal setting and task design, which can be achieved through SOPs [10][12] - A well-structured SOP can prevent application failures and ensure that the AI model provides relevant and coherent responses [12][48] - The article suggests that AI should have specific objectives to enhance the quality of interactions, making conversations more engaging and meaningful [20][48] Group 3: User Behavior and Expectations - Users often prefer concise answers, while others may seek more exploratory responses, creating challenges for AI in meeting diverse expectations [9] - The article warns against the misconception that AI can seamlessly handle multi-turn dialogues without proper design and structure [9][48] - It emphasizes the need for AI to adopt a more human-like approach in conversations, which can be achieved by embedding goals and emotional context into interactions [48][49]